Konrad Bühlmeyer

Compression of intrapulmonary bronchi by abnormally branching pulmonary arteries associated with absent pulmonary valves

In 3 patients with absent pulmonary valve syndrome and absent ductus arteriosus, the lungs were injected and analyzed postmortem using morphometric techniques. Two patients had tetralogy of Fallot and 1 had D-transposition of the great arteries, the latter being the first autopsy-proved case of absent pulmonary valve with transposition. In addition to the expected dilatation of the central pulmonary arteries and compression of the mainstem bronchi, postmortem pulmonary arteriography revealed a bizarre pattern of hilar branching. Instead of single segmental arteries, tufts of arteries arose which entwined and compressed the intrapulmonary bronchi. In all 3 patients the histologic structure of the pulmonary arteries was abnormal. The elastic lamina of the media of the right and left pulmonary arteries were increased in number outside the lung, but were decreased within the lung. At both sites, the elastic laminae were thickened and fragmented. In the 2 ventilator-dependent patients, there was slight medial hypertrophy and extension of muscle into normally nonmuscular arteries. In 1 of the 2 cases in which the number of bronchial generations was counted, they were decreased, and in the 1 case in which bronchial count was unknown, alveolar multiplication was severely impaired. Therefore, our data may explain why, in some patients with absent pulmonary valve syndrome, relief of compression of the mainstem bronchi alone does not appreciably alleviate or reverse severe respiratory disease.

Development of the pulmonary circulation in ventricular septal defect: A quantitative structural study

The application of quantitative morphometric techniques to evaluation of the lungs of nine children who died with a ventricular septal defect between the ages of 3 months and 4 years showed that the presence of pulmonary hypertension interferes with the growth and development of the pulmonary circulation. In all cases the preacinar arteries were of normal size and not dilated, and arterial size and number within the acinus were reduced and similar to those seen in the normal child at birth. Arterial and venous muscularity was increased as judged by an increase in wall thickness and by the presence of muscle in smaller and more peripheral arteries than is normal. Elevation of resistance was associated with failure of the intraacinar pulmonary circulation to develop normally rather than to obliterative pulmonary vascular disease. In view of the rapidity with which impairment of growth and elevation of resistance can develop, closure of a large defect is recommended before age 2 years.

Noninvasive assessment of coarctation of the aorta: Comparative measurements by two-dimensional echocardiography, magnetic resonance, and angiography

SummaryFifteen patients, aged between 9 and 21 years (mean, 15.1), with native coarctation of the aorta (CoA) or suspected recoarctation after surgical repair, underwent three different diagnostic procedures. Two-dimensional echocardiography (2D echo) and magnetic resonance imaging (MRI) of the thoracic aorta were performed in all patients; 14 patients underwent aortography, and digital subtraction angiography of the aorta was performed in one (after injection via a central venous catheter). Conventional electrocardiographic (ECG) gated MRI was performed, using the sagittal plane, a 256×256 acquisition matrix, multi-slice technique and a slice thickness of 10 mm. Diameters at the coarctation site were determined by all methods. Additional diameters of the descending aorta and the aortic arch were measured by MRI and echocardiography, respectively. All noninvasively obtained diameters were compared with angiographic data. Ultrasound imaging of the aortic isthmus was achieved in seven of 15 patients and of the aortic arch in nine of 15. The mean difference compared with angiographically determined diameters was 1.7 (0–7) mm, being greater for the coarctation site [mean, 2.2 (0–4)]. MRI images of the aortic isthmus were obtained in all patients, but the difference to angiographically determined diameters was slightly higher [mean, 3.2 mm (0–8)] than the ultrasound results. This deviation was presumably due to technical conditions, such as slice thickness and orthogonal imaging planes. Including all diameters, the correlation to invasive measurements wasr=0.82 (SEM=3.1) for MRI andr=0.89 (SEM=2.3) for echo recordings. It is concluded that MRI gives a reliable estimate of the severity of coarctation in a higher percentage of investigated patients. However, in patients in whom ultrasound imaging is successful, results are as reliable as those obtained by MRI.

Left ventricular myocardial mass determined by cross-sectional echocardiography in normal newborns, infants, and children

SummaryChanges in left ventricular muscle mass may be an important diagnostic or prognostic finding in children with congenital heart defects, but there are no data on normal mass as determined by cross-sectional echocardiography (CSE) in children. Fourteen newborns, 12 infants, and 69 children (1.5–17 years old) with a structurally normal heart were studied. Enddiastolic and end-systolic volume and mass were calculated from biplane-paired echocardiographic imaging planes—apical two-chamber and apical four-chamber views—using both an area/length and a Simpsons rule geometric method. Data were compared with M-mode measurements. There was a good correlation between area/length and Simpsons rule method [r=0.94, standard error (SE) 4 g/m2].To validate the measurements, interobserver data were gathered and end-diastolic and endsystolic mass measurements were compared. Interobserver variability for the measurements on the echocardiographic recordings was low at 4.8%; for end-diastolic mass the correlation between two observers wasr=0.99 (SE 3.3 g/m2) and for end-systolic massr=0.97 (SE 7.6 g/m2). Correlation between end-systolic and end-diastolic mass was acceptable (r=0.88, SE 5.9 g/m2) for the CSE mass determination, but poor for the M-mode measurements (r=0.51, SE 20.2 g/m2). Similarly, correlation between M-mode mass and mass estimated by CSE was poor, atr=0.58 for end-diastolic andr=0.094 for end-systolic mass. In newborns and infants the ratio of end-diastolic mass to end-diastolic volume was higher than in the older children, mainly because of a smaller ventricular volume in relation to body surface area in this age group.In the 26 newborns and infants, left ventricular mass was 48.7±10 g/m2 and volume was 34.6±7 ml/m2, yielding a mas to volume ratio of 1.41±0.03. For the 69 children (mean age 7.4±3.8 years), left ventricular mass was 63±10 g/m2 and volume 59±10 ml/m2 with a mass to volume ratio of 1.07±0.2. Left ventricular end-diastolic mass and volume strongly correlated with age (r=0.9) and body surface area (r=0.98 and 0.97, respectively). As CSE is easy to perform and repeatable, these normal values should provide a valuable database for further longitudinal studies of the development of left ventricular mass in patients with various cardiac abnormalities.

Comparison of magnetic resonance imaging with cross-sectional echocardiography in the assessment of left ventricular mass in children without heart disease and in aortic isthmic coarctation☆

Although left ventricular (LV) mass may be important to judge effects of left-sided cardiac obstruction or hypertension, reproducible noninvasively determined normal data in the pediatric age group are scarce. To validate cross-sectional echocardiographic LV mass determination, our data were compared with LV mass assessed by magnetic resonance imaging (MRI). MRI was considered to be a good reference method because there is usually no problem in defining endo- and epicardial borders with MRI. LV mass was assessed in 14 children aged 5.3 years (10 days to 14.7 years) with a mean body surface area of 0.78 m2 (range 0.25 to 1.61). With cross-sectional echocardiography the epicardial and endocardial volumes were calculated using a Simpsons rule algorithm in the apical 2- and 4-chamber view. The difference between epi- and endocardial volumes was multiplied by 1.05 to yield the mass. Mass was assessed with MRI using a multislice technique; the area of each myocardial slice was calculated and multiplied with the slice thickness, and the resultant slice volumes were added to obtain the myocardial volume. On cross-sectional echocardiography, the mass was 55 g (range 12 to 126) or 64 g/m2 (range 46 to 79); on MRI it was 60 g (range 33 to 87) or 69 g/m2 (range 46 to 89). Regression analysis yielded an r value of 0.98 with a standard error of the estimate of 5.7 g or a 10% difference. In older children, LV mass determined by MRI was bigger than the one derived by echocardiography. It is concluded that cross-sectional echocardiography can reliably assess LV myocardial mass in pediatric patients.

Pulmonary Atresia with Intact Ventricular Septum and Right Ventricle—Coronary Artery Fistulae: Selection of Patients for Surgery

The anatomy of the coronary arteries and fistulae, filling pattern and evolution, right ventricular (RV) anatomy (as demonstrated by cineangiocardiography), hemodynamics, and various surgical procedures were evaluated with respect to their effect on coronary artery blood supply to provide guidelines for surgical management and candidate selection.

Hypoplasia of the intrapulmonary arteries in children with right ventricular outflow tract obstruction, ventricular septal defect, and major aortopulmonary collateral arteries.

SummaryPostmortem injection studies have been carried out on the pulmonary vasculature of four children dying with pulmonary atresia and ventricular septal defect or severe tetralogy of Fallot with major aortopulmonary collateral arteries, in which nearly all bronchopulmonary segments had more than one source of blood supply. Despite regional variations in the source of blood supply, there was remarkable uniformity of arterial size and number within the respiratory unit throughout each case. In all cases, there was a normal number of arterial pathways, but both pre- and intraacinar arteries were considerably smaller than normal. The need for early operative intervention to ensure growth of pre- and particularly intraacinar arteries is emphasized.

Left ventricular function and myocardial mass after aortic valvotomy in infancy

SummaryEighteen of 25 survivors of aortic valvotomy in infancy were reinvestigated by cross-sectional echocardiography a mean of 7.5 (2.3–13.4) years after surgery. They had been operated at a median age of 38 (5–330) days. At the follow-up examination the gradient across the aortic valve was 41±19 (15–85) mmHg and the ejection fraction was 0.73±0.10 (0.48–0.84). Left ventricular (LV) end-diastolic volume was 66±17 (33–191) ml/m2. LV mass was 96±36 (44–204) g/m2 and the LV mass volume index (LVMVI) (mass divided by end-diastolic volume) was 1.43±0.4 (0.9–2.28). Eleven of 18 patients had an abnormally high mass volume index compared with 95 age-matched controls with structurally normal hearts. The correlation between the residual pressure gradient across the aortic valve and mass volume index yielded anr value of 0.75 (p<0.0004). One patient had been reoperated and underwent resection of a subaortic stenosis 4 years after the initial operation. Four patients with a resting gradient of more than 50 mmHg and one with grade 4 aortic regurgitation are scheduled for further surgical treatment.We conclude that, although LV function was normal in most patients who underwent aortic valvotomy in infancy, LV mass remains elevated in a significant number of patients, who may remain at risk of developing subendocardial ischemia.

Open heart surgery without homologous blood transfusion and subsequent use of recombinant erythropoietin to treat postoperative anemia

The purpose of our study was to assess the effect of administration of erythropoietin on postoperative anemia in patients with congenital heart disease who had undergone open heart surgery without the use of homologous blood. One intravenous and one subcutaneous dose of 300 units of erythropoietin per kg body weight was given to 10 patients at a mean of five (3–8) days after surgery. Another group of 10 patients, who had undergone open heart surgery but were not given erythropoietin postoperatively, were used as controls. There was a significant fall in all patients in hemoglobin from 13.6±1.3 gm/dl prior to surgery to 9.8±1.6 gm/dl after surgery. In patients treated with erythropoietin, there was an immediate significant rise in hemoglobin and hematocrit. Hemoglobin rose by 14% on the seventh day after treatment (the twelfth day after surgery) and by 25% on the tenth day after treatment (the fifteenth day after surgery). It had increased by only 8% 15 days after surgery in the controls. These findings suggest a beneficial effect of administration of recombinant erythropoietin on postoperative anemia in children who had undergone open heart surgery without the use of homologous blood.

DIAGNOSIS OF CONGENITAL HEART DEFECTS BY MEANS OF TRANSTHCRACIC 3-DIMENSIONAL ECHOCARDIOGRAPHY WITH A COMPUTED TOMOGRAPHY IMAGING PROBE (ECHO-CT)

A prototype tomographic ultrasound system: Echo-CT was evaluated 71 patients aged 4 days to 17 years with normal heart (n=1), ventricular septal delect (n=14), subaortic stenosis (n=10), mitral valve anomalies (n = 9), atrioventricular septal defect (n = 3), acrtic stenosis (n = 5), atrial septal defect (n=4) and various other congenital heart defects (n=20). The transducer is a 5 Mhz device with G4 elements mounted on a sliding carriage. The transducer is driven by a computer-controlled stepper motor in steps of .5-1.3 mms and acquires parallel tomographic slices of the head perpendicular to the position of the probe. The transducer moves from the outflow tract to the cardiac apex. With ECG and respiration gating a complete cardiac cycle is recorded at each tomographic level. 50 to 130 slices per patient wore acquired. Position of the probe on the chest varied in each patient. In 7 newborns the probe was used from the subcostal position. Image acquisition took 3-7 minutes. Data arc stored as three-dimensional datasels in the imaging computer. Out of those data sets up to 5 different two-dimensional views in different planes can be constructed, so that the heart can be viewed from multiple planes without changing the probe position on the chest. In all but 2 patients wo could also reconstruct the heart three-dimensionally from different views, which look 20-90 minutes per patient. Views similar to the ones a surgeon has at operation can be genenerated. We conclude that this preliminary study demonstrates the feasibility of the tomographic three-dimensional reconstruction technique, which yields additional information on the morphology of septal defects, tricuspid, mitral and aortic valves and complex malformations.