The role of lung biopsy in children with perinatally acquired AIDS.

There is a need for a rapid and efficacious method of diagnosis of pulmonary infiltrates in perinatal HIV infection. However, controversy still exists about which method--open lung biopsy (OLB) versus bronchoscopic techniques--is the best for this population. We present our results with OLB in 24 children with HIV-related lung disease. Over a 6-year period, 27 OLBs were performed on 24 children with diagnosis of HIV infection. The procedures were performed under general anesthesia using a limited anterolateral thoracotomy. Suspicious areas were removed with the autostapler. The specimens were studied for the presence of non-infectious as well as bacterial, viral, fungal, and mycobacterial diseases. There were no operative deaths related to the procedure. Morbidity was limited to prolonged but self-resolving air leaks in two patients (8.3%). Five hospital deaths occurred between 3 and 12 weeks postoperatively and 11 late deaths between 3 months and 6.5 years. All deaths were related to AIDS. Eight patients (33.3%) are still alive 2 to 8 years postoperatively. A total of 43 pathologies were found in 27 specimens. A positive pathologic finding was obtained in all patients, with two patients having nonspecific minimal changes. This resulted in a change of therapy in all but one case. The technique of OLB in children with AIDS is safe and simple. It should be performed early in the course of the disease and, a careful selection of candidates can minimize the incidence of complications.

Role of extended aortoplasty related to the definition of coarctation of the aorta.

One hundred thirty-nine patients underwent operation for coarctation of the aorta. Age ranged from 1 day to 21 years and weight, from 1.5 to 70.4 kg. Numerous methods of repair were used. The operative mortality was low (1.3%), and 17 patients (11.3%) died late. Recoarctation occurred in 13 patients (9.4%). We attempted to correlate mortality and recoarctation with the surgical procedure. A review of the literature revealed no classifications of coarctation that applied to the anatomical and pathological variations we found at the time of operation. Therefore, we devised a surgical classification to separate the various entities in the spectrum of coarctation: type I = primary coarctation; type II = coarctation with isthmus hypoplasia; and type III = coarctation with tubular hypoplasia involving the isthmus and segment between the left carotid and left subclavian arteries. Each of these types has subtypes: A = with ventricular septal defect and B = with other major cardiac defects. We believe that rather than labeling one procedure as "the procedure of choice," providing this classification will allow the surgeon to use a method of repair that is suited to the anatomical variation.

Neonatal aortic stenosis.

Aortic stenosis in the neonate has been associated in the past with a high operative mortality. As a result, in the current era of percutaneous balloon dilatation, the optimal mode of therapy remains controversial. An approach of stabilization with cardiopulmonary bypass, followed by relief of left ventricular outflow tract obstruction, was used at three institutions, and the results are presented. During the period 1983 to 1989, 40 neonates with isolated aortic stenosis and patent ductus arteriosus or coarctation of the aorta, or both, underwent operative therapy. Ages ranged from 1 to 30 days, median of 12 days, including 17 patients in the first week of life. There were 30 boys and 10 girls; weights ranged from 2.5 to 5.5 kg with a mean of 3.6 kg. Perioperative conditions included congestive heart failure in 38 and mitral regurgitation in 16; left ventricular-aortic gradients ranged from 15 to 130 mm Hg, with a mean of 67 mm Hg. There were 30 open valvotomies and 10 transventricular dilatations. The hospital survival rate was 87.5% (35/40) with no significant difference between the methods of valvotomy (9/10 in the transventricular dilatation group, 90%; 26/30 in the open valvotomy group, 87%). Although multiple methods of perfusion and valvotomy were used, the single unifying factor of cardiopulmonary bypass stabilization was present in all 40 patients. No significant difference in survival was noted between institutions, methods of cardiopulmonary bypass, cardiopulmonary bypass times, crossclamp times, or method of valvotomy. There have been five reoperations, with one late death in a patient requiring mitral valve replacement and an apical-aortic conduit. One sudden death occurred; autopsy revealed endocardial fibroelastosis. Results demonstrate that in the three institutions using the methods described, a high operative and late survival rate is possible. The results of this technique, against which percutaneous dilatation should be compared, are standard in the current era.

Experience with the polytetrafluoroethylene surgical membrane for pericardial closure in operations for congenital cardiac defects.

From 1984 through 1987, pericardial closure for 96 patients who underwent repair of congenital heart defects was completed with placement of a polytetrafluorethylene surgical membrane. Ages ranged from 1 day to 20 years (median age 2.1 years), and weight ranged from 2.7 to 44.6 kg (median weight 10.7 kg). Use of the membrane increased over the 4-year period. The patients were divided into three groups according to probability of reoperation: group I, low probability (35 patients); group II, intermediate probability (41 patients); and group III, high probability (20 patients). Early postoperatively there were no infections in any group and no complications attributable to the polytetrafluoroethylene membrane. No increase in pericardial drainage was found when the studied patients were compared to similar groups without the membrane. There were a total of 17 deaths (17.7%): group I, 1/35 (2.8%); group II, 9/41 (21.9%); and group III, 7/20 (35%). None of the deaths could be attributed to placement of the polytetrafluorethylene membrane. Seven patients had a second operation, which made possible the in vivo evaluation of the membrane. In addition, there was one late death, which allowed an autopsy. Findings were similar in all eight. There were no adhesions between the chest wall and the membrane. The latter had basically become transparent and was loosely attached to the epicardium. On the epicardial surface there was a thin layer of collagenous fibrous tissue that did not interfere with the visualization of the heart anatomy, including the coronary vessels. At early postoperative follow-up, difficulty in evaluation of the patient by two-dimensional echocardiography was encountered only in the parasternal view, but not in the other scans. In the parasternal view, the membrane is between the echo transducer and the heart and there is moderate blurring of finer details of the cardiac architecture. At late postoperative ultrasound study, resolution of the images in the parasternal views is much better. The following conclusions were reached: (1) Polytetrafluorethylene surgical membrane can be used for pericardial closure; (2) reoperation shows no adhesions to the chest wall or epicardium, and visualization of the cardiac surface, especially of the coronary vessels, is not obscured; (3) the membrane causes some difficulty in the echocardiographic examination of patients in the parasternal views, but other views allow for a complete postoperative evaluation of the heart.

Systemic-pulmonary polytetrafluoroethylene shunts in palliative operations for congenital heart disease. Revival of the central shunt.

The concept of central shunting in smaller children with the Waterston shunt was initially well accepted. It has been abandoned because of the difficult estimation of lumen size, preferential flow to the right side, and difficulty in the take-down of the shunt. We have replaced the Waterston shunt with a short segment of polytetrafluoroethylene between the ascending aorta and the main pulmonary artery. From January 1979 to December 1986, 190 shunt operations were performed in 157 patients, with the use of 26 classic Blalock-Taussig shunts (13.7%), six Waterston shunts (3.1%), nine Glenn shunts (4.7%), 80 central aortopulmonary polytetrafluoroethylene shunts (42.2%), and 69 modified Blalock-Taussig shunts (36.3%). Polytetrafluoroethylene grafts were used for 149 of the 190 (78.4%) shunts. Overall mortality was 15.2%, with nine early deaths (4.7%) and 20 late deaths (10.5%). Deaths were due to the complex nature of the congenital anomaly or definitive surgical repair. The patients weighed from 1.6 to 48 kg and ages ranged from 1 day to 22 years. We have modified our technique so that (1) graft length is less than 0.5 cm and both ends are beveled, (2) the aortotomy is fashioned with a punch, (3) the center of the polytetrafluoroethylene graft is never clamped, (4) heparin is given during the construction of the shunt, and (5) aspirin (10 mg/kg/day) is administered daily. Patency ranges from 1 to 4 years. We conclude that the polytetrafluoroethylene shunt provides excellent palliation and that the central shunt, in the smaller child and infant, offers the benefits of shunting without distortion of the peripheral pulmonary arteries.