Transmediastinal repair of complex coarctation and interrupted aortic arch.

In the past 2 years, eight patients have had repair of coarctation of the aorta or interrupted aortic arch along with an intracardiac procedure performed through a simple midsternotomy. Five underwent total repair and three had palliation for the intracardiac lesion. Four patients did well (including one who required a resection for recoarctation), two died intraoperatively, and two died postoperatively (11 and 21 days), one of them from clostridial sepsis and the other from cerebral anoxia that followed a cardiac arrest precipitated by a massive hemoptysis. We found that with this approach the aortic obstruction could be readily relieved and the intracardiac procedure carried out. Since the majority of these patients have compromised hemodynamic status, the ease of instituting immediate bypass, avoidance of a separate incision, and feasibility of total repair are major advantages.

Vascular anomalies causing tracheoesophageal compression. Review of experience in children.

Two hundred four infants and children (mean age 13 months) have undergone operation for the relief of tracheoesophageal obstruction resulting from vascular anomalies. One hundred thirteen patients had complete vascular rings (group I), 61 with double aortic arch and 52 with right aortic arch with a left ligamentum. Nine patients had a pulmonary artery sling (group II), 71 had innominate artery compression (group III), and 11 had miscellaneous anomalies (group IV). Patients were admitted with respiratory distress, stridor, apnea, dysphagia, or recurrent respiratory infections. Diagnosis was established by barium esophagogram in group I; barium esophagogram, bronchoscopy, and computed tomography or angiography in group II; bronchoscopy in group III; and barium esophagogram or angiography in group IV. The operative approach was through a left thoracotomy in group I, II and IV (93% of these patients) and through a right thoracotomy for group III (96% of these patients). The operative mortality rate was 4.9% and there were seven late deaths (3.4%). There have been no operative deaths in patients with isolated vascular anomalies in the past 28 years. Follow-up data from 1 month to 20 years (mean 8.5 months) were available on 159 patients; 141 (92%) were essentially free of symptoms, and 12 (8%) had residual respiratory problems. Five of six patients in group II having a lung scan postoperatively had a patent left pulmonary artery. A strong index of suspicion is necessary to avoid the complications of vascular rings in children. Barium swallow is the best single diagnostic technique for patients with complete vascular rings. A bronchoscopic study is required to diagnose innominate artery compression. Angiograms or computed tomographic scans are used to confirm the diagnosis of pulmonary artery sling. Left thoracotomy provides excellent exposure for all vascular rings except the displaced innominate artery, for which a right thoracotomy is the best approach.

Development of neo-coarctation in patients with transposed great arteries and hypoplastic aortic arch after Lecompte modification of anatomical correction.

We report on three patients with kinking in the proximal aortic arch that developed after Lecompte modification of the arterial switch operation. Two patients had a previous subclavian patch repair of coarctation of the aorta and had an associated hypoplasia of the transverse aortic arch, and one patient had hemodynamically mild coarctation at the anatomical repair. A severe pressure gradient across the kinked area ("neo-coarctation") necessitating reoperation developed in one patient. The acute arch angulation appears to be due to an excessive posterior displacement of the ascending aorta by the anterior relocation of either the right or left main pulmonary artery branch from underneath the aortic arch. A foreshortened and frequently hypoplastic transverse aortic arch, a common association with coarctation of the aorta, appears to be especially vulnerable to the development of "neo-coarctation" after the Lecompte modification of the anatomical repair of transposed great arteries.

Extended aortic valvuloplasty for recurrent valvular stenosis and regurgitation in children.

Recurrent significant aortic valvular stenosis or regurgitation, or both, after balloon or open valvotomy in pediatric patients often necessitates aortic valve replacement. In an attempt to preserve the aortic valve, we performed extended aortic valvuloplasty in 21 children with recurrent aortic valve stenosis or regurgitation from January 1989 to March 1993. Previous related procedures were one open aortic valvotomy or more (n = 15), balloon valvotomy (n = 4), balloon valvotomy after surgical valvotomy (n = 1), and repair of iatrogenic valve tear (n = 1). Mean age at the time of the extended aortic valvuloplasty was 6 +/- 3.4 years. Mean pressure gradient across the aortic valve was 56 +/- 12 torr. Regurgitation was moderate (grade 2 to 3) in nine and severe (grade 4) in 12 patients. Extended aortic valvuloplasty techniques consisted of thinning of valve leaflets (n = 15), augmentation of scarred and retracted leaflets with autologous pericardium (n = 11), resuspension of the augmented leaflet (n = 14), release of the rudimentary commissure from the aortic wall (n = 5), extension of the valvotomy incision into the aortic wall on both sides of the commissure (n = 20), patch repair of the sinus of Valsalva perforation (n = 1), reapproximation of tears (n = 5), and narrowing of the ventriculoaortic junction (n = 2). No operative deaths occurred. The postoperative mean pressure gradient, assessed by most recent Doppler echocardiography or cardiac catheterization at a follow-up of 18 +/- 6 months, was 19 +/- 6 torr (p < 0.01 versus the preoperative gradient). Aortic regurgitation was absent in 13, mild in 6, and moderate-to-severe, necessitating subsequent aortic valve replacement, in 2. This short-term experience indicates that extended aortic valvuloplasty is a safe and effective surgical approach that minimizes the need for aortic valve replacement in children with significant recurrent aortic valve stenosis or regurgitation.

Modified Blalock-Taussig shunt in newborn infants.

The modified Blalock-Taussig shunt, interposing an expanded polytetrafluoroethylene graft between the subclavian and pulmonary arteries, was performed in 30 neonates with a mean age of 8.8 days and a mean weight of 3.14 kg. Underlying lesions included severe tetralogy of Fallot or its variant (N = 10), transposition complex (with pulmonary stenosis or atresia) (N = 6), single ventricle equivalents (with pulmonary atresia or stenosis) (N = 9), and pulmonary atresia with intact ventricular septum (N = 5). The mean preoperative arterial oxygen tension prior to prostaglandin E1 therapy was 29.5 torr. The shunt was performed through a right thoracotomy in 18 patients, through a left thoracotomy in nine, and through a median sternotomy in three. A 5 mm graft was used in 21 patients and a 6 mm graft in nine patients. The mean postoperative arterial oxygen tension was 64.1 torr (p less than 0.001). The incidence of early shunt occlusion was 3.3% and the hospital mortality was 3.3%. Actuarial functional life of the shunt (no death or reoperation related to shunt failure) was 91% at 3 years' follow-up. Nine patients were recatheterized. There was no distortion of the pulmonary artery. The ratios of the diameter of the right pulmonary artery and pulmonary valve anulus to that of the descending aorta increased after the operation by 50% (p less than 0.001) and 52% (p less than 0.05), respectively. Our experience indicates that the modified Blalock-Taussig shunt has an excellent function, offers several technical advantages, and lacks most of the drawbacks of other systemic-pulmonary artery shunts. It may be the shunt of choice in patients less than 1 month of age.

Tracheoplasty with pericardial patch for extensive tracheal stenosis in infants and children.

Five infants with long tracheal stenosis were operated upon by means of a pericardial patch tracheoplasty. The approach was through a median sternotomy with extracorporeal circulation for respiratory support. In four, the obstruction was due to complete rings; in the other, there was an associated tracheal trauma which had occurred during resuscitation. This patient requires prolonged stenting with a tracheostomy tube. All others are asymptomatic postoperatively, with the longest follow-up being 22 months. There were no deaths or infections. We conclude from this experience that median sternotomy provides an excellent approach to the trachea, that autogenous pericardium is advantageous, and that there is no need for prolonged tracheal stenting in most patients.

The surgical management of left ventricular outflow tract obstruction due to tricuspid valve pouch in complete transposition of the great arteries.

Subpulmonary stenosis in transposition of the great arteries, resulting from a tricuspid valve pouch bulging into the left ventricular outflow tract through a ventricular septal defect, can be missed at the time of operation in the flaccid, nonbeating heart unless preoperative diagnosis has been established. In our experience, six patients were found to have this lesion. In four patients the tricuspid valve pouch was recognized preoperatively. At operation, retraction of the tricuspid valve pouch into the right ventricle, patch closure of the ventricular septal defect, and a Mustard procedure were performed in three patients; the fourth is awaiting correction following initial palliation with a subclavian-pulmonary shunt. In the other two, the ventricular septal defect was partially or completely obliterated by a tricuspid valve pouch that was missed preoperatively and during exploration at the time of the Mustard procedure. Residual left ventricular outflow tract obstruction was subsequently corrected with a left ventricle-pulmonary artery valved conduit. Echocardiographic and angiocardiographic examinations offer helpful signs for the diagnosis of tricuspid valve pouch. Transatrial retraction of the redundant tricuspid valve tissue into the right ventricle, patch closure of the ventricular septal defect, and Mustard operation are the procedures of choice. A left ventricle-pulmonary artery valved conduit may be required for residual unresectable left ventricular outflow tract obstruction.

Tetralogy of Fallot with absent pulmonary valve. Should valve insertion be part of the intracardiac repair?

Thirteen patients with complex of tetralogy of Fallot and absent pulmonary valve were treated. Three infants presented with severe cardiorespiratory distress (Group I) necessitating surgical intervention in two and intensive medical treatment in the third; there were no late survivals. The remaining 10 patients were operated upon electively at an older age (Group II); in six the ventricular septal defect was closed and the right ventricular outflow obstruction relieved (Group IIA), and in the other four a valve was inserted at the pulmonary anulus at the time of intracardiac repair (Group IIB). There was no hospital mortality in Group II. Five of the six patients in Group IIA remained symptomatic postoperatively, and three needed a subsequent pulmonary valve insertion. All patients in Group IIB were symptom free following operation; one patient began having a progressive decrease in exercise performance, decrease in pulmonary artery size, and cardiothoracic ratio were more marked in those patients who had primary insertion of a pulmonary valve. The clinical results and postoperative assessment of cardiac function tend to indicate that pulmonary valve insertion should be performed at the time of the intracardiac repair.

Surgical approach to severely symptomatic newborn infants with tetralogy of Fallot and absent pulmonary valve.

The surgical management of severely symptomatic newborn infants with tetralogy of Fallot and absent pulmonary valve has been controversial, and the results of a variety of operative approaches have not been satisfactory. We report on a technique for the treatment of these patients, which consists of (1) ligation of the main pulmonary artery to eliminate pulmonary regurgitation, excessive right ventricular stroke output, and secondary pulmonary artery dilation and airway obstruction and (2) insertion of a subclavian-pulmonary artery polytetrafluoroethylene shunt to provide pulmonary blood flow. The procedure was used in four neonates. Two patients operated on at 2 and 3 days of age are doing well 15 and 19 months postoperatively. The other two, operated on at 3 and 4 weeks of age after unsuccessful prolonged medical treatment and positive-pressure ventilation, failed to show long-term improvement and died of sepsis and respiratory failure 3 and 5 months after operation. This experience, though limited, suggests that early surgical intervention to control pulmonary regurgitation prevents progressive pulmonary artery dilatation and secondary bronchial compression, decreases the need for prolonged preoperative and postoperative ventilation, and improves the outcome of these critically ill neonates.

Transposition of the great arteries with intact ventricular septum. Arterial switch in the first month of life.

Twenty-three infants with simple transposition of the great arteries and intact ventricular septum were operated on from October 1983 to October 1986. The age at operation in 22 infants ranged from 2 to 21 days and in one was 35 days (mean 9.82 +/- 6.86 days). The infants were evaluated with cardiac catheterization at 1 to 27 days of age. Twenty-two infants had balloon atrial septostomy, and 22 received prostaglandin E1 infusion. The left ventricular diastolic wall thickness, assessed by M-mode echocardiograms, varied between 2.8 and 4 mm. There were two hospital deaths in this group of 23 infants (mortality 8.6%), and there were no late deaths. All surviving patients are doing well clinically. One patient had asymptomatic nonsustained ventricular tachycardia necessitating phenytoin. Postoperative echocardiographic assessment performed on 15 patients at 0.93 +/- 0.61 years of age and cardiac catheterization and angiographic studies on seven patients at 1.07 +/- 0.13 years after operation revealed excellent ventricular performance, good semilunar valve function, and mild gradient at the right ventricular outflow with a mean right ventricular pressure of 37.4 +/- 4.1 torr.

Arterial switch in simple and complex transposition of the great arteries.

Arterial switch for repair of transposition of the great arteries was performed on 53 patients since October 1983. These patients were divided into three groups: group I, 25 infants with an intact ventricular septum who had primary repair in the first month of life (2 to 34 days of age, mean 9.7 +/- 6.6); group II, 13 patients with an intact ventricular septum who had anatomic repair after a preliminary procedure (pulmonary artery banding in 13, shunt in 10, atrial septectomy in 1); and group III, 15 infants with transposition of the great arteries and ventricular septal defect. In group III, six patients had Taussig-Bing abnormality, nine had previous pulmonary artery banding, three had coarctation of the aorta repaired earlier in life, and four were less than 2 weeks old. Overall early mortality was 9.4% (5/53: group I 8%, group II 7.6%, group III 13.3%). Two late deaths occurred in group II 10 and 12 weeks postoperatively after infection and high fever. A third late death 18 weeks postoperatively was due to aspiration in an infant with Goldenhar's syndrome. Mortality and morbidity decreased significantly after an initial learning period (no deaths from July 1985 to March 1987 overall, and none in the last 15 infants operated on in group I). The surviving 45 patients are doing well. All have normal sinus rhythm. Two had transient asymptomatic arrhythmias. Left and right ventricular function assessed by echocardiogram and postoperative cardiac catheterization were within normal ranges in all but two patients, one with pulmonary artery stenosis and one (Taussig-Bing abnormality with two large ventricular septal defects) with severe pulmonary vascular disease (9.6 units) observed before anatomic repair. The right ventricular pressure at catheterization ranged from 27 to 42 mm Hg in 12 patients and was 55 mm Hg in two. There was no aortic stenosis. Aortic insufficiency was trivial in three patients and mild in one. We conclude that excellent results can be obtained with arterial switch for transposition of the great arteries with or without ventricular septal defect, especially in neonates.

Technical modifications for improved results in total anomalous pulmonary venous drainage.

To delineate factors that contribute to improved surgical outcome in patients with total anomalous pulmonary venous drainage, we reviewed the records of 52 consecutive patients. Venous drainage was supracardiac in 25 (48%), cardiac in 12 (23%), infracardiac in 10 (19%), and mixed in five (10%). Preoperative pulmonary venous obstruction was present in 18 patients (35%). Median age at the time of repair was 35 days and weight, 3.7 kg. Repair was performed with deep hypothermia, low-flow cardiopulmonary bypass, and occasional short periods of circulatory arrest. In patients with coronary sinus drainage, the veins were tunneled to the left atrium through an enlarged atrial septal defect, with a mortality of 8% (1/12) and no postoperative stenosis. The approach in patients with supracardiac, infracardiac, and mixed drainage varied with time. In 16 patients, the condition was managed by apical or right-sided exposure of the common vein, anastomosis of the common vein to the left atrium with continuous sutures, and primary closure of the atrial septal defect (type I repair). In the other 24 patients the common vein was approached from the right side through the right atrium and the interatrial septum. Common vein-left atrium anastomosis was performed with interrupted sutures and a piece of pericardium used to augment the anastomosis, prevent common vein distortion, and close the atrial septal defect (type II repair). Mortality in type I repair was 25% (4/16) and in type II repair, 4% (1/24). Follow-up was 7.86 +/- 4.0 years with no late deaths. Postoperative stenosis occurred in five of 14 (36%) patients who had type I repair versus two of 23 (9%) who had type II repair. Multivariate analysis showed that type I repair was a positive risk factor for hospital mortality (p = 0.05) and restenosis (p = 0.04). The technique of transatrial exposure of the common venous chamber, interrupted suturing of the common vein to the left atrium, and pericardial patch augmentation significantly improves survival and decreases risk of restenosis.

Surgical management of occluded conduits after the Fontan operation in patients with Glenn shunts.

Five complete conduit occlusions occurred in four patients with the Glenn shunt 2 months to 2 years after the Fontan operation. The possible reasons for complete conduit occlusion were severe dehydration, high pulmonary vascular resistance, and intraoperative manipulation of the conduit. In one patient in whom complete conduit occlusion developed twice, no possible cause could be identified. Surgical approaches included replacement of the occluded conduit in three patients and creation of an atrial septal defect and left aortopulmonary shunt in the fourth patient. All patients who had replacement of the occluded conduit survived. The fourth patient had severe cyanosis and hypoxemia from marked reduction of flow through the Glenn shunt because of reversal of flow through large venous collaterals. He subsequently died of Candida sepsis. A fifth patient (previously reported) who had complete conduit occlusion also died after a similar procedure. We believe that in patients with a Glenn shunt who develop complete conduit occlusion after the Fontan operation, conduit excision and a secondary Fontan operation, preferably without the use of woven Dacron, should be done instead of establishing an atrial septal defect and aortopulmonary shunt to the left lung.

The role of the Glenn shunt in patients undergoing the Fontan operation.

Twenty-seven patients (3 to 22 years) underwent the Fontan operation. Seventeen had tricuspid atresia and 10 had other complex lesions. Sixteen patients had a Glenn shunt (12 prior to, three simultaneous with, and one following the Fontan operation). One early death (3.7%) occurred in a 4 1/2-year-old child with tricuspid atresia II-C, previous pulmonary artery banding, and a closing ventricular septal defect (80 mm Hg gradient). There were two late deaths (7.4%) from Candida sepsis, after 4 and 6 months, respectively. The 24 patients who survived the Fontan operation had postoperative hospital stays of 6 to 90 days (average 18). Patients with tricuspid atresia and an established Glenn shunt (nine patients, Group I) had postoperative hospital stays of 7 to 19 days (average 9.5), and none had significant pleural or pericardial effusions. Patients with tricuspid atresia without a Glenn shunt (seven patients, Group II) had postoperative hospital stays of 6 to 60 days (average 17.5), with three having significant effusions. Of the patients with other complex lesions, all without an established Glenn shunt, five had significant effusions. Four additional major complications (two tricuspid patch disruptions with ineffective pulmonary blood flow and two complete occlusions of a valved conduit) were encountered in which the Glenn shunt proved lifesaving. We believe that an established Glenn shunt played a major role in attaining minimal postoperative hemodynamic instability, effusions, renal failure, and mortality in our patients. The Glenn shunt should be considered in patients who are less than ideal candidates for the Fontan operation.

Unusual mitral valve abnormalities complicating surgical repair of endocardial cushion defects.

A review of 155 cases of surgically repaired endocardial cushion defects revealed 16 patients (10%) with additional unusual mitral valve abnormalities that complicated the surgical procedure. Eight patients had accessory mitral valve tissue that connected the anterior and posterior leaflets to form a double-orifice valve (Group I). In four (50%), the lesion was associated with intermediate atrioventricular canal and small left ventricle; all four died following repair. In the other four, it was associated with ostium primum defect; all survived and are well. A single papillary muscle in the left ventricle was present in six patients (Group II). Two had intermediate atrioventricular canal and both died postoperatively. The other four had complete endocardial cushion defect and three are well following the operation. Perforation of the valve leaflets was present in two patients with ostium primum (Group III). Both patients are well postoperatively. Modification of the surgical technique is required to effect satisfactory repair. The bridge connecting the posterior and anterior leaflets of the mitral valve should be left undisturbed. Otherwise, severe regurgitation may result. In patients with single papillary muscle and complete atrioventricular canal, repair may be accomplished by borrowing from the tricuspid portion of the anterior leaflet, rotating that part posteriorly, and partially closing the cleft. Small perforations of the mitral leaflet do not require closure and do not result in regurgitation. Echocardiographic and angiographic delineation of these abnormalities and thorough intraoperative exploration are important in avoiding pitfalls at the time of repair.

Preparation of the left ventricle for anatomical correction in patients with simple transposition of the great arteries. Surgical guidelines.

Pulmonary artery banding in combination with an aortopulmonary shunt was performed on 16 patients with simple transposition of the great arteries to prepare the left ventricle for anatomical correction. Three groups were identified after operation: Group I (four patients) had increased pulmonary blood flow and tight pulmonary artery banding; Group II (four patients) had increased pulmonary blood flow and moderate pulmonary artery banding; Group III (eight patients) had normal pulmonary blood flow and moderate pulmonary artery banding. Postoperative low cardiac output was present in all patients in Group I, whereas mild heart failure was present in two patients in Group II and in two in Group III. There was one hospital death (6%). The follow-up period was 125 patient-months. Left ventricular systolic pressure rose from 63 +/- 11 torr before the operation to 101 +/- 35 torr after the procedure in Group I (p less than 0.05), from 59 +/- 10 to 93 +/- 33 torr in Group II (p less than 0.05), and from 55 +/- 10 to 84 +/- 16 torr in Group III (p less than 0.005). The increase in left ventricular muscle mass was from 44 +/- 2 gm/m2 preoperatively to 108 +/- 12 gm/m2 after operation in Group I (p less than 0.01), from 43 +/- 3 to 93 +/- 8 gm/m2 in Group II (p less than 0.02), and from 46 +/- 3 to 55 +/- 14 gm/m2 in Group III (p = no statistically significant difference). The postoperative change in left ventricular end-diastolic volume was from 100% +/- 17% to 133% +/- 23% of normal in Groups I and II (p less than 0.05) and from 123% +/- 29% to 107% +/- 36% of normal in Group III (p = no statistically significant difference). In preparing the left ventricle for anatomical correction, avoidance of severe pulmonary artery banding decreases the incidence of postoperative myocardial dysfunction, a moderate degree of volume overload and pulmonary artery banding provides the most effective stimulus for ventricular growth, and a small to moderate atrial septal defect is advantageous because it ensures the volume preload necessary for the development of the left ventricle.

Factors that exaggerate the deleterious effects of pulmonary insufficiency on the right ventricle after tetralogy repair. Surgical implications.

Postoperative cardiac catheterization data of 74 patients with pulmonary insufficiency after tetralogy repair were analyzed. Two groups were identified: Group A, 26 patients with normal right ventricular function (ejection fraction 95% +/- 5.5%, end-systolic volume 110% +/- 17% of predicted normal) and Group B, 48 patients with right ventricular dysfunction (ejection fraction 80% +/- 18% [p less than 0.001], and end-systolic volume 218% +/- 75% of predicted normal [p less than 0.001]). There was no significant difference between the two groups with respect to frequency of previous palliative procedures, age at operative repair, operative techniques, methods of myocardial protection, and follow-up period. Right ventricular dysfunction in Group B was associated with significant distal pulmonary stenosis (right ventricle-pulmonary artery pressure gradient 28 +/- 13 torr in Group A versus 55 +/- 20 torr in Group B, p less than 0.001), moderate pulmonary regurgitation (regurgitant fraction 18% +/- 11% in Group A versus 32% +/- 10% in Group B, p less than 0.001), and large transannular outflow patch (ratio of patch diameter to descending aorta diameter 1.31 +/- 0.16 in Group A versus 2.50 +/- 0.28 in Group B, p less than 0.001). Pulmonary valve insertion was performed in 42 patients in Group B. Eighteen had subsequent cardiac catheterization. Right ventricular function recovered completely (end-systolic volume 122% +/- 24%, and ejection fraction 92% +/- 7% of predicted) in five of six patients (83%) who had valve insertion within the first 2 years after tetralogy repair. In contrast, right ventricular function remained abnormal in all 12 patients who had valve insertion later than 2 years after tetralogy repair (p less than 0.05). Patients with residual pulmonary stenosis and/or a large transannular outflow patch are at risk for the development of right ventricular dysfunction from pulmonary insufficiency after tetralogy repair. Early correction of these residual lesions and control of pulmonary insufficiency may prevent long-term deterioration in right ventricular function.

Experience with St. Jude Medical valve prosthesis in children. A word of caution regarding right-sided placement.

Thirty-six children aged 6 months to 18 years, underwent insertion of 37 St. Jude Medical cardiac prostheses. In 20, the valve was placed in the aortic or mitral position, and in 16 in the pulmonary or tricuspid position. There was one (2.8%) hospital death. All patients received maintenance doses of salicylates and dipyridamole after the operation. Follow-up data are available for all patients for 12 to 24 postoperative months. There was no incidence of valve dysfunction or thromboembolic complication in any of the 20 patients with valves in the systemic (left) side of the circulation, and all manifested improvement in their functional class. In contrast, six (37%) of the 16 patients with valves in the pulmonary (right) side of the circulation developed dysfunction of the prosthesis 1 to 6 months after insertion. Prosthesis failure was associated with fibrous tissue growing into the struts, leading to leaflet immobilization. At 2 years, the actuarial functional life was 100% for mitral and aortic valves and 70% for pulmonary and tricuspid valves. The data illustrate the excellent hemodynamic function of the St. Jude Medical valve in children. The absence of thromboembolic complications warrant continued implantation of the prosthesis in the left side without warfarin anticoagulation therapy, but the high incidence of valve dysfunction in the pulmonary position does not justify its continued use in the right side.

Fontan type operation for complex lesions. Surgical considerations to improve survival.

Twenty-five of 49 patients who underwent a Fontan type operation had complex lesions other than tricuspid atresia with ventriculoarterial concordance. Three patients had significant subaortic stenosis. Thirty-four palliative operations, including nine Glenn shunts, were performed before the Fontan operation. Direct atriopulmonary anastomosis was performed in 21 patients. In four, valved conduits were used. Twelve patients had right atrioventricular valve patch closure (three had running and nine had interrupted suture technique). On the basis of the presence of increased or decreased pulmonary blood flow before any surgical intervention, patients were divided into Group 1 (previous pulmonary artery banding, N = 8) and Group II (pulmonic stenosis, N = 17). Postoperatively, in Group I, 87% had significant effusions, mean right atrial pressure was higher (20.6 +/- 6.5 torr), and hospital stay longer (31 days). In Group II, 40% had significant effusions, mean right atrial pressure was lower (16.5 +/- 4.3 torr), and hospital stay shorter (15 days). Significant atrioventricular valve patch disruption occurred in three patients (two had running suture technique), and conduit occlusion occurred in two. Four patients (three with subaortic stenosis and pulmonary artery banding) without an established Glenn shunt required Fontan takedown for persistent low cardiac output, two of whom died (2/25 or 8%). There were three late deaths (3/23 or 13%). Nineteen of 20 surviving patients observed from 2 months to 6 years are doing well. We believe that early Fontan takedown in patients with persistent low cardiac output, interrupted suture technique for atrioventricular valve closure, avoidance of valved conduits, and a preliminary Glenn shunt in patients with pulmonary artery banding and/or subaortic stenosis can further improve the results with the Fontan operation for complex lesions.

Persistent low cardiac output after the Fontan operation. Should takedown be considered?

Four of 44 patients who had undergone the Fontan operation had persistent low cardiac output necessitating takedown of the shunt 6 to 65 hours (average 23 hours) postoperatively. All four were in a group of 22 patients with complex lesions other than tricuspid atresia with ventriculoarterial concordance. The development of postoperative right atrial hypertension (average 24 torr), hepatomegaly, marked ascites, and decreasing lung compliance led to severe systemic hypotension with systolic arterial pressure ranging from 55 to 82 torr (average 68 torr), persistent metabolic acidosis, and oliguria despite massive colloid and crystalloid infusions (11,000 ml/m2/24 hr) and inotropic support. At reoperation the atriopulmonary anastomosis, which was found to be wide open, was taken down and an atrial septal defect was created in all patients. Three patients were left with a Glenn shunt and an aortopulmonary shunt to the left lung. One patient had bilateral aortopulmonary shunts. Two patients who survived reoperation had immediate postoperative improvement in systolic arterial and mean right atrial pressure (average 100 torr and 11.5 torr, respectively). Both are well 5 months and 4 years later. Repeat Fontan operation remains a possibility with acceptable risks because of the presence of the Glenn shunt in both patients. We believe that takedown should be considered in patients with persistent low cardiac output after the Fontan operation.