Ten years have passed since the publication of our article “Aortic allografts: reconstruction of right ventricle-pulmonary artery continuity.” Our initial experience demonstrated the utility of such an approach in patients ranging from 6 weeks to 26 years, and conduit sizes from 11 to 26 mm. Specific points included the low risk in infants with primary repairs, such as those with truncus arteriosus, and in conduit replacements in older patients; the decreased incidence of hemorrhage, especially in the infant group with tissue to tissue anastomosis; the possibility of decreased infectious potential with tissue reconstruction; that a larger conduit size might be placed without compression; conduit survival might be extended; and pseudointimal proliferation and valve calcification, the mechanism of Dacron heterograft failure, might be avoided. Ten years of use has provided new insights and the influence of type of allograft and size, age of patient, as well as the unique immunologic/degenerative potential appreciated. In the first of these, the type of allograft, contemporary with the publication of our manuscript, we and others began the use of pulmonary allografts, most specifically in the neonatal population. Several large series failed to correlate type of allograft with structural failure [1Cleveland D.C. Williams W.G. Razzouk A.J. et al.Failure of cryopreserved homograft valved conduits in the pulmonary circulation.Circulation. 1992; 86: 150-153Google Scholar, 2Stark J. Cull C. Stajevic M. Jothi M. Elliott M. de Leval M. Fate of subpulmonary homograft conduits determinants of late homograft failure.J Thorac Cardiovasc Surg. 1998; 115: 506-516Abstract Full Text Full Text PDF PubMed Scopus (161) Google Scholar], but multiple others, including the CryoLife pooled data, demonstrated increased structural deterioration (in the latter at 7 years aortic 65% vs pulmonary 93%, p < 0.018). The effect of size of allograft has been increased by the use of smaller conduit in smaller infants and repair of complex congenital cardiac lesions in the neonate with lower risks and complications, a major impact of allograft availability. Thus, although Tam and associates confirmed that in a given patient, larger conduits have a greater longevity (23 patients with truncus, mean age 35 days, mean size 15.6 mm, mean time of replacement 8 years 2 months) [3Tam R.K. Tolan M.J. Zamvar V.Y. et al.Use of larger sized aortic homograft conduits in right ventricular outflow tract reconstruction.J Heart Valve Dis. 1995; 4: 660-664PubMed Google Scholar], smaller rather larger conduits afford the potential for early repair, and it is in this subset that the maximal difference in aortic and pulmonary allograft results is observed. Age of patient clearly influences the development of the immunologic/degenerative response with calcification of the arterial wall and both narrowing and shortening of the allograft, intensifying the degree of stenosis [4Bando K. Gordon G.K. Schaff H.V. Mair D.D. Julsrud P.R. Puga F.J. Outcome of pulmonary and aortic homografts for right ventricular outflow tract reconstruction.J Thorac Cardiovasc Surg. 1995; 109: 509-518Abstract Full Text Full Text PDF PubMed Scopus (186) Google Scholar]. As identified in our original manuscript, the use of allografts provide significant technical improvements in tissue-to-tissue anastomosis and the avoidance of compression and hemorrhagic complications. However, unique immunologic/degenerative changes occur in such grafts, most markedly in the small aortic allograft in the young individual. An intense debate continues as to the etiology of these changes [5Yankah A.C. Yacoub M.H. Hetzer R. Cardiac Valve Allografts. Steinkopff Darmstadt, Darmstadt, Germany1997Google Scholar, 6Rajani B. Mee R.B. Ratliff N.B. Evidence for rejection of homograft cardiac valves in infants.J Thorac Cardiovasc Surg. 1998; 115: 111-117Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar, 7Mitchell R.N. Jonas R.A. Schoen F.J. Pathology of explanted cryopreserved allograft heart valves comparison with aortic valves from orthotopic heart transplants.J Thorac Cardiovasc Surg. 1998; 115: 118-127Abstract Full Text Full Text PDF PubMed Scopus (156) Google Scholar, 8Kawauchi M. Nakajima J. Takeda M. Oka T. Takamoto S. Aortic valves are antigenic but less so than myocardium.J Thorac Cardiovasc Surg. 1998; 116: 532Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar]. Considering the nature of the cryopreserved allograft with preservation of cellular elements, it now seems amazing that a privileged position in the vasculature and “fewer conduit replacements” was considered. Data by Stark and associates suggests that although the mechanism of conduit obstruction has changed (calcification and stenosis vs pseudointimal proliferation and valve degeneration), longevity may not have improved [3Tam R.K. Tolan M.J. Zamvar V.Y. et al.Use of larger sized aortic homograft conduits in right ventricular outflow tract reconstruction.J Heart Valve Dis. 1995; 4: 660-664PubMed Google Scholar]. The viable allograft, however, presents us with the unique opportunity to intervene and alter conduit survival unavailable in the Dacron heterograft prothesis. Elucidation of the mechanism of structural failure may provide therapeutic options to improve this “conduit of choice” [5Yankah A.C. Yacoub M.H. Hetzer R. Cardiac Valve Allografts. Steinkopff Darmstadt, Darmstadt, Germany1997Google Scholar, 8Kawauchi M. Nakajima J. Takeda M. Oka T. Takamoto S. Aortic valves are antigenic but less so than myocardium.J Thorac Cardiovasc Surg. 1998; 116: 532Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar].

The 12 mm Dacron conduit containing a porcine valve is the smallest valved conduit manufactured and is used in the youngest infants with the most diminutive pulmonary arterial system. The outcome of patients with such a conduit is unknown. Between 1975 and 1985 there were 49 hospital survivors after placement of a 12 mm extracardiac valved conduit from the right ventricle to the pulmonary artery. Follow-up is available in 42 patients, aged 1 to 16 months (mean 3.5) and weighing 2.5 to 8.7 kg (mean 3.8). Twenty-eight patients (67%) have undergone subsequent conduit replacement, and 11 (26%) are alive and asymptomatic with a mean follow-up of 56 months. There were three late deaths. The interval between implantation and conduit change was 4.5 to 101 months (mean 44), allowing a weight gain of 2.7 to 23 kg (mean 10.4) before reoperation at age 12 to 117 months (mean 49). Despite elevated right ventricular pressures equaling systemic values, 37% of these patients were clinically asymptomatic. The gradient across the 12 mm valved conduit before explantation ranged from 30 to 173 torr (mean 83) with an almost equal predilection for stenosis at the proximal anastomosis, valve, conduit, distal anastomosis, and main pulmonary artery. The intervening pulmonary artery growth determined the size of the replacement conduit, 14 to 25 mm (mean 16), and was the main factor influencing the results of reoperation. This study demonstrates that the 12 mm porcine valve-containing conduit affords palliation in this difficult subset of patients with the smallest pulmonary arterial tree.

Surgical treatment of transposition of the great arteries and intact ventricular septum has changed markedly in the past 10 years. However, long-term follow-up on new approaches is not available. In a unique group of patients, operated on in the first 100 days of life with the Mustard procedure, such follow-up is available, and the results of these true infant repairs represent a benchmark against which subsequent techniques applied to infants should be compared. During the period of 1975 to 1980, 36 infants, less than 100 days of age, who had transposition of the great arteries and intact ventricular septum, underwent Mustard repair at the University of California, San Francisco. Ages ranged from 4 to 98 days (mean 46 days) and weights from 2.3 to 6.6 kg (mean 3.5 kg). There were no early deaths, and late follow-up was available from 8 to 13 years (mean 10 years). The late survival rate was 97% (mean 10 years). There was a 62% rhythm disturbance-free survival rate, 89% reoperation-free survival rate, and 91% pacemaker-free survival rate. Echocardiographic evaluation revealed obstruction of the superior vena cava in eight patients, tricuspid insufficiency in four, right ventricular dysfunction in two, and left ventricular outflow tract obstruction in three. The Mustard procedure, performed in the first 100 days of life, results in a high rate of survival (early 100% and late 97%, at a mean of 10 years) and a low incidence of late complications against which other techniques of infant repair should be compared.

To evaluate the potential usefulness for characterization of tissue and anatomical changes associated with cardiac transplantation rejection by nuclear magnetic resonance imaging (MRI), sixteen dogs underwent heterotropic cardiac transplantation with six not immunosuppressed serving as controls. Myocardial biopsy and MRI were obtained and compared on a weekly basis. Untreated allografts showed a significant increase in T2 and intensity values by MRI compared to the native heart as early as one week after transplantation. The MRI findings corresponded to the histological progression of acute rejection process in both treated and untreated groups. The linear relationship between histology and MRI was 0.72 while the correlation between T2 and the water content was 0.92. Serial gated MRI correlated with chronic anatomical changes of transplant rejection with evidence of progressive or increasing myocardial wall thickness and decrease in ventricular chamber size.

The surgical experience in 75 patients with total anomalous pulmonary venous connection (TAPVC) between 1975 and 1986 was reviewed. Most of these patients underwent operation at less than 1 month of age (39 of 75, 52%). Operative approaches used were the standardized left-sided approach for supracardiac and infracardiac TAPVC, reserving transatrial repair for only the intracardiac lesions. Cardiopulmonary bypass was managed by deep hypothermia and circulatory arrest in 16 of 75 (29%) and continuous hypothermic bypass with low flow was used for the remaining 59 of 75 (71%). Operative policy included maintaining patency of existing patent ductus (1 patient) and atrial (4 patients) or ventricular (3 patients) defects to allow additional shunting in cases of severe obstruction. Operative mortality was 9.3% (7 of 75) and late mortality was 6.7% (5 of 75). The determinants of operative deaths were (1) the type of TAPVC (highest rates in types III and IV with 3 of 17 or 19%, and 2 of 8 or 25%, respectively) and (2) associated severe preoperative pulmonary-to-systemic venous obstruction. Determinants of late death were (1) residual pulmonary-to-systemic obstruction (two patients type III and 1 patient type I), and (2) associated intracardiac lesions (two patients type IV with atrioventricular canal). In summary, the timing of corrective surgery and outcome are influenced by the clinical condition of the infant as a direct reflection of the anatomic type and severity of associated lesions.

A variety of techniques have been described for the repair of aortic coarctation. Most of these use operative exposure through a left thoracotomy. There are, however, instances when a median sternotomy provides a more versatile approach and allows for equally acceptable repair. Twelve patients (age, 2 days to 16 years) with coarctation of the aorta serve to illustrate the value of the technique. All but 3 patients were under 3 years old. Five patients had repair of coarctation combined with repair of other congenital cardiac anomalies (atrial and ventricular septal defects, 2 patients; atrial septal defects, 2 patients; and valvular aortic stenosis, 1 patient). Six patients had complicated recurrences of previously repaired coarctations and 1 patient had a primary coarctation that involved the aortic arch. All repairs were accomplished through a median sternotomy with the use of cardiopulmonary bypass and periods of (1) total circulatory arrest (6 patients); (2) reduced flow (4 patients); or (3) normal flow with multiple sites of arterial inflow (2 patients). All patients did well with no operative mortalities and no sequelae to date (8 weeks to 4 years). Examples of various techniques are provided to demonstrate the flexibility obtainable with this method. Median sternotomy can provide a viable approach to difficult problems involving coarctation of the aorta.

From 1981 to 1987, 62 patients with ventricular tachyarrhythmias and associated sudden death required treatment after unsuccessful initial medical therapy (51 patients) and previous surgical therapy (11 patients). Surgical options included direct revascularization (group I, 11 patients), endocardial resection (group II, 7 patients), automatic internal cardiac defibrillators (group III, 18 patients), and these combinations of operations: revascularization and endocardial resection (group IV, 18 patients), revascularization and insertion of automatic internal cardiac defibrillators (group V, 5 patients), and endocardial resection and insertion of cardiac defibrillator (group VI, 3 patients). Five underwent repeat revascularization (4) or endocardial resection (4) with mitral valve replacement (1) or papillary muscle reconstruction (2). The overall operative mortality of 8.1% (5/62) has been acceptable (2.8% or 1/36 for the simple procedures [groups I, II, and III] and 15.4% or 4/26 for the combined procedures [IV, V, and VI]; p less than 0.074). Operative risk factors included recent myocardial infarction (4/5 deaths) and depressed ejection fraction of 23% or less (5/5 deaths). These operative risks were highest in group IV and, thus, the highest mortality was in group IV (4/18 patients or 22%, p less than 0.022). Six late deaths (4 patients in group III) brought the overall survival rate to 82% or 51/62 patients at a mean follow-up of 30 months. Surgical treatment of sudden death ventricular tachyarrhythmias requires a planned, combined operative approach, since initial medical or surgical failures can occur. The optimal surgical approach requires complete revascularization, resection of the localized subendocardial arrhythmogenic focus, and ventricular reconstruction. An automatic defibrillator is a palliative alternative for less favorable surgical anatomy and for patients who are poor operative candidates for these combined or reoperative procedures.

Between 1975 and 1985, 125 infants 2 to 365 days old (majority, 30 days old or less) with coarctation of the aorta underwent surgical repair. Forty-seven patients (38%) had severe congestive heart failure (CHF), metabolic acidosis, and poor systemic perfusion. The predominant operative technique was synthetic patch aortoplasty (100 patients); the remaining 25 had an end-to-end anastomosis. There were no operative deaths. Perioperative complications were minimized with the synthetic patch technique (less than 15%). For patients surviving at least 3 months after repair, the arm-leg systolic blood pressure gradient was relieved in 82% (71/87) of the patients having patch aortoplasty versus 65% (15/23) of the patients with end-to-end anastomosis. Although the rate of reoperation between the two groups was similar (patch, 5 [6%]; end-to-end, 3 [13%], two of the reoperations in the patch group were for preexisting hypoplastic transverse aortic arch. Late deaths (20 patients, 16%) were due to other major associated cardiac anomalies. Patch aneurysms have not occurred. Expedient use of synthetic patch aortoplasty has decreased perioperative complications, relieved coarctation gradients for CHF, increased early survival even in the presence of complex or associated cardiac anomalies, and has an acceptable rate of recurrent coarctation (6 to 13%).

An original surgical procedure had been developed entailing the implantation of a right-sided heterotrophic cardiac graft in 14 dogs. The benchwork preparation included creation of both atrial and ventricular septal defects which allowed transplantation with the avoidance of extracorporeal circulation and the utilization of all transplanted myocardium and parts. The structure and function of this graft were sequentially evaluated using rapid cine-computed tomograms which permitted assessment of forward graft flow, ventricular wall contraction, and diastolic thickness for both the native and grafted hearts. These parameters were followed during the early (E) (less than 3 days), intermediate (I) (4-21 days), and late (L) (greater than 21 days) postoperative periods. Forward flow (E: 3.0, I: 1.5, L: 2.6) through the transplant was maintained by the competence of the implanted valves despite a progressive decrease in the contraction (E: 3.0, I: 1.3, L: 1.0) and associated decrease in the wall thickness (E: 3.0, I: 2.6, L: 2.4). These findings were consistent with severe cardiac graft rejection without immunosuppression. In summary, long-term structural and functional forward flow of this configurated right-sided transplantation had been maintained by the competent valves on both sides of the graft despite severe rejection without immunosuppression as documented by rapid cine-computed tomograms.

The optimal method of anticoagulation in children with mechanical heart valves is controversial. Between 1975 and 1986, aspirin or aspirin with dipyridamole has been used for anticoagulation in children receiving a mechanical aortic valve at the University of California, San Francisco. Fifty-one patients (ages 1 to 23 years, mean 12.9 years) were treated with aspirin (n = 45) or aspirin with dipyridamole (n = 6) and observed a mean of 36.5 months (range 3 to 100 months). There were four late deaths: two from endocarditis and two from other medical problems, but none related to thrombosis or embolus. Follow-up was accomplished by direct contact with the patient, parent, or referring physician. Two patients (3.9%) were lost to late follow-up. One minor neurologic event occurred perioperatively and resolved spontaneously. There were no postoperative thromboembolic events. Eleven asymptomatic children were recently studied by magnetic resonance imaging or computed axial tomography of the brain and had no evidence of prior silent cerebral thromboembolic defects. There were four patients (5.9%) who had minor hemorrhagic complications: Three patients had nosebleeds and one patient had an upper gastrointestinal hemorrhage. Five patients were changed to warfarin anticoagulation: the patient with upper gastrointestinal hemorrhage and four older patients because of physician preference, all after uncomplicated aspirin therapy. There were no mechanical valve failures, although one patient required reoperation 9 months later for perivalvular leak. All children have remained in normal sinus or paced rhythm during follow-up. These results show that children with mechanical aortic valves in normal sinus rhythm can be safely treated with aspirin (or aspirin with dipyridamole) with little risk of thromboembolic events, valve thrombosis, or valve failure. Hemorrhagic complications resulting from aspirin are minor and easily treated.

Platelet deposition in the coronary microvasculature has not been completely defined in the temporal relationship to acute myocardial ischemia, the application of crystalloid cardioplegia, and during reperfusion on heart bypass. Twenty-two canine hearts were serially biopsied for the analysis of radioactively tagged platelets. Eleven hearts underwent an isolated heart support preparation with seven followed by potassium cardioplegic arrest and reperfusion while the remaining 4 were maintained on continuous bypass. All 11 hearts undergoing bypass experienced transient (less than 90 sec) ischemia during bypass preparation and produced platelet aggregation in the myocardium (51.12 +/- 24.0 as compared to nonischemic control group 12.3 +/- 4.7; P = 0.005). Potassium cardioplegia did not completely wash out these platelets to the nonischemic control levels (27.8 +/- 14.9; P = 0.04). With the onset of reperfusion after 1 hr of cardioplegic arrest, platelet radioactivity profoundly increased (133.3 +/- 72.8; P = 0.0101) and remained high throughout the hour of reperfusion (324.7 +/- 269.3; P = 0.0369). In summary, intracoronary platelets are activated after transient ischemic episodes during initiation of heart bypass. These ischemia-activated platelet aggregations persist despite the application of cardioplegia during the arrest period. This deposition, in turn, allowed an ongoing pattern of platelet aggregation during the early and subsequent reperfusion. This pattern of ischemia-activated platelet aggregations probably accounts for the progressive reperfusion injury and support of an antiplatelet treatment for coronary microvasculature protection.

The antithrombotic effects of prostacyclin infusion on myocardial platelet deposition were studied in a canine model during and after global ischemia. Eleven isolated heart preparations were subjected to 1 hour of cardioplegic arrest under moderate hypothermia (27 degrees to 28 degrees C), including a control group (n = 7) and a prostacyclin-treated group (n = 4). The hearts of four other dogs were continuously perfused for 180 minutes. Platelet deposition was measured at 15 minute intervals throughout the 3 hour study. Serial full-thickness myocardial biopsy specimens were analyzed for activity of 111In-labeled platelets with 99mTc-labeled erythrocyte correction for tissue blood content. The pattern of platelet distribution was determined by scintiscans of each heart, taken with a gamma camera at the end of the 60 minute reperfusion period. Substantial myocardial platelet deposition was found in the control hearts after ischemia but not in the prostacyclin-treated group (p less than 0.05). Furthermore, prostacyclin infusion had a significant disaggregatory effect on intracoronary platelet deposits when the precardioplegic and postcardioplegic biopsy specimens were analyzed (p less than 0.05). Three hours of continuous perfusion did not increase tissue 111In-labeled platelet activity. Ex vivo images showed platelet deposition to be a diffuse patchy process with significantly more 111In activity in the endocardium than in the epicardium after global ischemia (p less than 0.05). These data show the potent antithrombotic properties of prostacyclin in preventing and disaggregating ischemia-induced intracoronary platelet deposition during and after cardioplegic arrest.

To determine whether prostacyclin (PGI2) plays a beneficial role in the blood-perfused heart undergoing global ischemia, 20 isolated canine hearts were studied after sustaining one hour of cardioplegic arrest under moderate hypothermia (27°C to 28°C). Left ventricular function (peak systolic pressure, rate of rise of left ventricular pressure [dP/dt], and compliance change in left ventricular volume), myocardial edema, coronary blood flow, and oxygen content were measured during the preischemic period and at 15 and 30 minutes during reperfusion.

The oxygen (pO2) and carbon dioxide (pCO2) gas tensions of a pneumonectomy space were studied in an animal model. The 2 gases were measured in a standard blood gas analyser. The space gas pO2 and pCO2 equilibrated to a steady state within 48 hours of thoracotomy (pO2 mean 53.9 +/- 9.3 SD torr; pCO2 mean 44.8 +/- 9.9 SD torr). In the presence of a bronchopleural fistula (BPF), the pO2 rose significantly (mean pO2 118.4 +/- 13.2 SD torr; p less than 0.001). The space pCO2 usually fell, but the response was inconsistent (mean pCO2 33.6 +/- 16.4 SD; p = NS). A further rise in space pO2 could sometimes be induced by allowing the animals to breathe 100% oxygen by face mask. Data from 4 clinical cases of BPF support the concept that measurement f the space gas pO2 could be used as a bedside test for BPF.

The authors discuss the early management of patients with transposition of the great arteries in the neonate. The success of early operative procedures in the neonatal period has limited the usefulness of the Rashkind balloon septostomy. The authors propose that both the Senning and arterial switch procedure may be better accomplished if the septostomy is not performed. The Senning operation appears to be the current method of choice and early operative intervention is highly successful.