- Email: [email protected]
Congenital Cardiac Anomalies
Congenital Cardiac Anomalies One-Stage Repair in Infancy Donald B. Doty, M.D., Ronald M. Lauer, M.D., and J. L. Ehrenhaft, M.D. ABSTRACT A proposed preferred treatment plan consisting of one-stage operative repair of congenital cardiac anomalies whenever complete correction was feasible and operative intervention required was tested during the period February 1,1972, to September 15,1974. Experience with 74 patients aged 14 hours to 24 months allows some conclusionsto be made regardingthe advisability of this approach. For patients with ventricular septa1 defect and tetralogy of Fallot, this approach has quite acceptable results and certainly is more direct than palliative operations and second-stage repair. The hypothesis is more difficult to justify in patients with transposition of the great arteries. By avoiding repair in patients under 4 months of age and using a better technique for placement of the intraatrial partition, improved results are anticipated. In the miscellaneous group of anomalies, results depend upon the complexity of the defect and the ability to achieve accurate anatomical correction. After an early mortality of 20% and some late deaths, survival and improvement resulted in 51 of these babies. These results justify persistence in the choice of one-stage correction when operative intervention is required.
C
orrection of congenital cardiac anomalies has traditionally been performed in children when they reach the age of 5 or 6 years. By this time the child is generally of sufficient size that the technical aspects of operation are minimized and the deleterious effects of cardiopulmonary bypass can be distributed over a large body mass. If a child required operative intervention earlier in life, generally some form of palliative operation, such as banding of the pulmonary artery or systemic-to-pulmonary artery shunt, was performed because of the high mortality rates attending primary repair of these defects in infants and young children [4, 71. The results of palliative operation [ 15, 20, 231 and the problems attending the required second-stage operative repair [ 11 made this approach less than optimal. The advent of better blood oxygenators and blood filtering devices as well as the improved ability to manage cardiopulmonary bypass, often combined with hypothermia and circulatory arrest techniques, have allowed primary, one-stage repairs of a variety of congenital cardiac defects in infants. A number of highly successful series were reported in the period 1971-1973 [lo, 11, 13, 18,211,andthesubjecthasbeenreviewedbyRittenhouse [14] and Bonchek [3] and their associates. From the Divisions of Thoracic and Cardiovascular Surgery, Department of Surgery, and of Pediatric Cardiology, Department of Pediatrics, T h e University of Iowa Hospitals and Clinics, Iowa City, Iowa. Accepted for publication Feb. 27, 1975. Address reprint requests to Dr. Doty, Division of Thoracic and Cardiovascular Surgery, Department of Surgery, The University of Iowa Hospitals and Clinics, Iowa City, Iowa. 52242. 316
T H E ANNALS OF THORACIC SURGERY
Congenital Cardiac Anomalies in Infancy Encouraged by the results of others, a program of intracardiac repair of congenital cardiac anomalies, regardless of age at the time of necessary operative intervention, was elected at The University of Iowa Hospitals. All patients in whom repair was feasible were accepted, and none was excluded because of associated anomalies or general condition. This report details the results in 74 consecutive patients in order to determine if early repair of congenital cardiac defects improves results.
Clinical Information and Methods There were 74 patients with a variety of congenital cardiac anomalies requiring operative intervention during the period February 1, 1972, to September 15, 1974 (see Table 1).The age of the patients ranged from 14 hours to 24 months, and they weighed 2.4 to 14.7 kg (Figure). Growth was retarded in most of the patients, with 61 falling below the sixteenth percentile for weight. All these patients had intracardiac repair of their defect. The only palliative operations performed during this period were done in patients in whom correction was not feasible. The indications for operation in patients with defects having predominantly left-to-right intracardiac shunts were intractable congestive cardiac failure or advancing pulmonary vascular resistance. Patients with anomalies producing arterial desaturation were operated upon when hypoxic spells occurred or if cyanosis worsened and polycythemia advanced. There were 20 patients with a hematocrit over 60% and 11 over 70%; the highest was 82%. Two techniques were used to manage the circulation during operation. The first group of 28 consecutive patients were operated on using deep hypothermia and circulatory arrest for the intracardiac repair with a limited period of extracorporeal circulation for cooling and rewarming. This standardized technique was utilized in order to collect data on the central nervous system consequences and has been reported elsewhere [5].After a low incidence of permanent nervous system damage was documented in this initial group, hypothermia and arrest techniques continued to be utilized with confidence in selected cases, especially in very small patients (less than 6 kg) and in those with complex intraatrial defects. A total of 51 patients had their circulation arrested to facilitate operative repair. There were, in addition, 23 patients in whom the circulation was continuously maintained by cardiopulmonary bypass. These were patients in whom the intracardiac repair could be expected to require less than 40 minutes, the exposure to repair the defect would be straightforward, and who were generally larger (weight greater than 6 kg). In these patients, circulatory arrest seemed to provide no great advantage over continuous bypass with moderate hypothermia; an initial 5-minute period with the perfusate at 15°C was followed by a period of aortic occlusion up to40 minutes, during which the perfusion temperature was 28°C and the flow to the patient 1.8 liters per minute per square meter of body surface area. Choice of perfusion technique or the use of circulatory arrest did not appear to affect mortality, as deaths were related to the cardiac anomaly and its repair rather than to the life-support technique used during the operation. Standard methods of intracardiac repair of defects were used in all patients VOL.
20, NO. 3, SEPTEMBER, 1975
317
DOTY, LAUER, AND EHRENHAm WEIGHT
Percentile 84 50 16
t
0
I
I
I
I
1 2 3 4
I
I
6 12 AGE (months)
I
18
1
24
Age and weight of 74 patients having early correction of congenital cardiac anomalies.Dots within circles repfesent early death, showing that most occurred inpatients less than 4 months old. Predicted weighfor age is superimposed by percentile to emphasize the marked growthfailure in these patients.
except for 14 with transposition of the great arteries who had the intraatrial baffle constructed of tubular weave-knit Dacron [8]. This has now been discarded in favor of pericardium. Elective operation was performed in 59 patients and emergent procedures in 15. The latter group is particularly noteworthy in that 12 subsequently died. Two of these patients had cardiorespiratory arrest prior to the procedure, and 1 underwent operation with a heart rate less than 40. Six of the patients had pulmonary edema and were on the respirator, one with a Pk0, of 114 mm Hg in spite of ventilatory assistance. Seven patients had a Pao, below 27 mm Hg, 1 of whom had seizures and a Pao, of 17 mm Hg. A second intracardiac operation was required in 5 patients, making a total of 79 procedures in these 74 patients. The follow-up period was 2 weeks to 30 months .
Results Since this is not a uniform group of patients, the results are analyzed according to the lesion. Table 1 presents the mortality occurring during the 30 days after operation. There were 15 deaths, which is 20% of the patients. Ten of the 15 early deaths occurred in babies less than 4 months old (see the Figure). The cause of death in 8 patients who died later is shown in Table 2. VENTRICULAR SEF'TAL DEFECT
Twelve patients had closure of ventricular septal defect because of intractable heart failure often associated with growth retardation. There were no hospital or late deaths. The age range was 17 days to 24 months with 5 patients less than 4 months old. Two patients had associated anomalous muscle bands in the right ventricle that obstructed the outflow and required excision, and 3 had multiple muscular septal defects. Nearly all the patients had high pulmonary blood flow, and pulmonary resistance ranged from 1.36 to 6.0 units. The ratio of pulmonary to systemic resistance was greater than 0.25 in 6 patients, the highest being 0.66. Two patients have small residual shunts(QJQ 1.16 to 1.3). One child has bron318
THE ANNALS OF THORACIC SURGERY
Congenital Cardiac Anomalies in Infancy TABLE 1. HOSPITAL MORTALITY FOLLOWING REPAIR OF CONGENITAL CARDIAC ANOMALIES IN PATIENTS 24 MONTHS OF AGE OR LESS
Anomaly Ventricular septal defect Tetralogy of Fallot With pulmonary atresia Transposition of great arteries Simple Complex Miscellaneous Double-outlet right ventricle Atrial septal defect Coarctation, VSD, ASD, PDA Total anomalous pulmonary veins Pulmonary stenosis Aortic stenosis Atrioventricular canal (complete) Hemitruncus Cor triatriatum Aortopulmonary window, VSD, anomalous coronary artery Total
No. of Patients
Mortality
< 30 Days 0
12 16 2
2
14 11
4 4
3
0 0 1
2 2
1
3 2 2 2 1 1 1
1 0 1 0 0 0
74
15
1
VSD = ventricular septal defect; ASD = atrial septal defect; PDA = patent ductus arteriosus.
chial asthma which is limiting development, and another has continued symptoms of left ventricular dysfunction. The others have done well in terms of general physical condition and exercise tolerance; growth improvement has been unpredictable. TETRALOGY OF FALLOT
The 18 patients in this group represented the full spectrum of the anomaly. The age range was 4 to 24 months with only 4 patients less than 12 months old. The indications for operation were hypoxic spells in 12 patients and increasing cyanosis and polycythemia in 6. Patients were selected for repair of the defect based on angiographic anatomy that suggested the feasibility of satisfactory anatomical correction. Generally, this meant pulmonary arteries of apparently “good’size and the pulmonary valve annulus measuring at least one-third of the aortic valve annular diameter. Only 2 patients had right ventricular hypoplasia and were classified as having pulmonary atresia because there was barely continuity between the right ventricle and pulmonary artery through the markedly hypoplastic pulmonary annulus. Since both patients died following patch reconstruction of the right ventricular outflow tract in addition to the usual tetralogy of Fallot repair, no further patients with unfavorable anatomy were operated on for repair. During this period a total of 9 aortopulmonary shunt procedures have been performed at this institution for palliation of patients with severe forms of tetralogy of Fallot or pulmonary atresia. There were 2 deaths (mortality, 22%). Seven of the 9 patients were less than 8 weeks old, and 5 operations were done in VOL. 20, NO. 3, SEPTEMBER, 1975
3 19
DOTY, LAUER, AND EHRENHAET TABLE 2. CAUSES OF 8 LATE DEATHS
Anomaly Transposition of great arteries Simple Complex Double-outlet right ventricle Atrial septal defect Atrioventricular canal (complete) Cor triatriatum
Months after Operation
6 14 1% 5
7 4
3
2
Cause of Death
.
Pulmonary venous obstruction Pulmonary venous obstruction Heart failure Superior vena cava obstruction Obstructive pulmonary vascular disease Sudden death, cause unknown Obstructive pulmonary vascular disease & mitral valve incompetence Pulmonary venous obstruction
the first week of life. Shunt procedures were generally favored in patients less than 4 months of age, and the anatomy was critically analyzed in all patients less than 12 months old when considering the choice of procedure when operative intervention was required, Using these rigid criteria in the selected 16 patients with favorable anatomy operated upon for repair of tetralogy of Fallot, only 1 died (mortality, 6%).Except for a patient who developed choreoathetosis after operation, all are doing remarkably well. TRANSPOSITION OF T H E GREAT ARTERIES
There were 25 patients 14 hours to 22 months of age in the group with transposition of the great arteries. Fourteen of the patients were less than 6 months old. A Rashkind balloon atrial septostomy was performed at the initial cardiac catheterization in 2 1 patients, a Blalock-Hanlon atrial septectomy was done at another institution in 1, and 3 patients had large ventricular septal defects so that mixing of pulmonary and systemic blood was thought to be adequate at initial catheterization.. Five patients did not improve following Rashkind atrial septostomy and underwent repair within 48 hours after cardiac catheterization. Of those improved by septostomy, the salutary clinical effect lasted from 1 to 2 1 months with an average of 7 months until surgical intervention was required. Fourteen had transposition of the great arteries with intact ventricular septum and no associated anomaly other than patent ductus arteriosus in 6. Transposition of the great arteries with associated infundibular pulmonary stenosis was present in 4, ventricular septal defect in 4, and both pulmonary stenosis and ventricular septal defect in 3 patients. Eight of the patients died less than 48 hours after operation (32%). Five of these were treated in the initial 4 months' experience. Four patients died 1 ?4to 14 months later (16%).A number of factors appeared to be associated with mortality. Age was very important, in that 9 of the 12 deaths were in patients who underwent operation at less than 4 months. All 4 patients having the intraatrial baffle operation (Mustard procedure) when Rashkind septostomy failed during the first week of life died, 2 of them shortly after operation and 1 at 5 weeks due to low 320
THE ANNALS OF THORACIC SURGERY
Congenital Cardiac Anomalies in Infancy cardiac output with apparently technically sound procedures; the fourth died at 14 months from pulmonary venous obstruction. Technical aspects of the operation and placement of the intraatrial partition to redirect venous return were important factors in 8 of the 12 deaths. Two patients had unrelieved obstructing hypertrophy of the outflow areas in the right or left ventricle. In 2 patients the pericardial baffle was redundant and appeared to obstruct pulmonary venous outflow. Another patient died of supraventricular tachyarrhythmia, perhaps related to the placement of the suture line relative to the conduction pathways. Three late deaths were due to pulmonary venous or superior vena cava obstruction, probably related to the configuration of a Dacron intraatrial conduit. One patient died of central nervous system complications 2 days after operation. Seizures had been marked prior to operation, and the hematocrit was 82% with a PaOz of 17 mm Hg. The convulsive disorder had been thought to be hypoxic in origin, but postmortem examination revealed sagittal sinus and cortical vein thromboses that had been present prior to operation. Reoperation for pulmonary venous or superior vena cava obstruction was performed in 4 patients, all of whom had Dacron intraatrial conduits. Two subsequently died and 2 are presently well. A good result was achieved in 13 patients with follow-up to 23 months. There was a remarkable improvement in general well-being and exercise tolerance as well as a dramatic change in the children’s appearance. MISCELLANEOUS ANOMALIES
The miscellaneous group includes 19 patients with a variety of cardiac anomalies. The experience with these patients is too limited for conclusions to be drawn regarding individual anomalies, but the group considered as a whole allows some general impressions. Complex anomalies and young babies seemed the rule, with an age range of 2 days to 22 months and 9 of the 19 under 4 months old. Double-Outlet Right Ventricle. Two patients with subaortic ventricular septal defect have done well while a third with indeterminate ventricular septal defect died 7 months after operation. This patient had advanced pulmonary arteriolar occlusive disease, marked desaturation, and polycythemia with a hematocrit of 72%. The pulmonary resistance of 9.3 units remained abnormal after operation, and progressive dilation of the right ventricle, especially the outflow portion, preceded death. Atrial Septal Defect. One patient had cardiorespiratory arrest and congestive cardiac failure prior to closure of the atrial septal defect and died suddenly 4 months after operation; no cause was identified. Another patient, operated on because of cardiomegaly and heart failure, has continued to have cardiac enlargement and mitral valve incompetence and had a presumed cerebral embolism with right hemiparesis 2 months after operation. The nervous system defect subsequently cleared, but the clinical course suggests myocardiopathy. Ventricular and Atrial Sepal Defect, Patent Ductus Arterww, and Coarctation of the Aorta. This difficult combination of anomalies has been repaired in a single VOL. 20, NO. 3, SEPTEMBER, 1975
321
DOTY, LAUER, AND EHRENHAFT operation using both sternotomy and lateral thoracotomy incisions in 2 patients. Encouraged by the first patient, who had an excellent result, the operation was attempted in a 1-month-old infant with respiratory insufficiency due to left-toright shunt. In spite of respiratory therapy, the Pacon continued to rise and was 114 mm Hg prior to operation. The patient died 24 hours after the procedure with what appeared to be low cardiac output and a hemorrhagic disorder due to disseminated intravascular coagulation. Total Anomalous Pulmonary Venous Connection. One patient with a subdiaphragmatic connection and another with connection to the coronary sinus did well, and the result was dramatic. Another patient with the supracardiac variety died because a patent ductus arteriosus was present which was not ligated. Pulmonary Valve Stenosis. Two patients with this anomaly did well; 1 of them had associated defects of the atrial and ventricular septa and patent ductus arteriosus. Aortic Valve Stenosis. One patient, 2 days old, was operated upon after a cardiac arrest and in spite of adequate valvulotomy died in the operating room. Postmortem examination showed severe endocardia1 fibroelastosis of the left ventricle. Another patient, 2 months old, did well. Complete Atrioventricular Canal. Two patients were successfully operated upon at 13 and 22 monthsof age, respectively. However, 1 of them, who had heart failure, cyanosis, and pulmonary vascular resistance of 8.8 units, did not improve. Subsequent cardiac catheterization showed systolic right ventricular pressure to be 110 mm Hg and left ventricular pressure 96 mm Hg. There was a right-to-left shunt (Q&& 0.65) and mitral valve incompetence. The child died 3 months postoperatively. Anomalous Origin of Right Pulmonary Arteryfrom Aorta. This anomaly (hemitruncus arteriosus) was successfully corrected in 1 patient by disconnecting the right pulmonary artery from the aorta and creating an anastomosis to the main pulmonary artery. Cor Triatriatum. Successful operation at 5 weeks in 1 patient with initial improvement was followed by progressive pulmonary venous obstruction. The patient died 2 months later at a second operation. Scarring of the pulmonary venous orifices and moderate pulmonary venous hypoplasia were found. Aortofmlmonary Window, Ventricular Septa1 Defect, and Anomalous Oripn of Coronary Arteriesfrom Pulmonary Arteries. The coronary artery anomaly was not detected prior to operation in 1 patient but was diagnosed and properly repaired by patch closure of the aortopulmonary window in such fashion as to bring the coronary arteries in continuity with the aorta through the window. Myocardial ischemia time, however, resulted in tetanic contracture of the ventricles that could not be reversed.
Comment The hypothesis that one-stage repair of congenital cardiac anomalies will be associated with improved results when operative intervention is required, regard322
T H E ANNALS OF THORACIC SURGERY
Congenital Cardiac Anomalies in Infancy less of age, has been tested. N o attempt was made to favor the results in this group of patients, as all were accepted regardless of general condition if complete repair was feasible. Perhaps the results would have been better if those with cardiac arrest, respiratory failure, or central nervous system damage had been eliminated from the study. By accepting all patients into the series, a more representative sample of the benefits of this approach was tested. Subgroup analysis allows some conclusions regarding the utility of primary correction of congenital heart defects in infancy. In the ventricular septal defect group, it is clear that early operation is safe and is preferable to palliative banding of the pulmonary artery. The uniform survival and good late results in our patients agree completely with the experience in a large group reported by Mori and associates [ 101. Even in patients with complex and multiple ventricular septal defects who undergo operative intervention early in life, good results can be anticipated. Similarly, tetralogy of Fallot can be repaired early with operative mortality rates as satisfactory as with the traditional two-stage approach. Certainly, postoperative respiratory management seems easier in patients who have had less time to develop extensive aortopulmonary collateral circulation. This experience is supported by the results of others [12, 17, 221, so early one-stage repair may be performed in these patients provided that the anatomy, especially the size of the pulmonary annulus, permits it. It should be noted, however, that judging the adequacy of the pulmonary annulus may be difficult in patients less than 4months old. This small experience does not justify extension of early repair to patients with severe hypoplasia of the right ventricular outflow tract or pulmonary atresia, as suggested hy Barratt-Boyes and Neutze 121. The overall results in our patients with transposition were not as favorable as in the experience of others [9, 191. A number of factors are probably responsible for the increased mortality, not the least of which was the technical conduct of the operation. Technical problems were associated with 8 of the 12 deaths, 5 of which occurred at the beginning of the series when understanding of these matters was less precise. Since we adopted Subramanian’s modification [19] of the Mustard operation, the results have clearly improved. This technique, using pericardium, takes a final configuration remarkably similar to the Brom modification, which Stark and colleagues [16] have used so successfully in small patients. The age of patients having the Mustard procedure is very important to the results achieved. As shown in this series, 8 of the 12 deaths occurred in patients less than 4 months old. Increased mortality has also been observed by Subramanian and Wagner [ 191 in patients with simple transposition of the great arteries, in patients less than 1 month old, and in patients less than 6 months old with associated complex anomalies. It should probably be predicted therefore that this series so heavily weighted with young babies (over half less than 6 months old) would have an increased mortality. The complex factors responsible for failure of the Rashkind procedure to achieve palliation in the first week of life apparently were still present following the intraatrial baffle operation, contributing to results so poor that this approach has been abandoned. Similar experience has directed Carpena VOL. 20, NO. 3, SEPTEMBER, 1975
323
DOTY, LAUER, AND EHRENHAFT and Subramanian [6] back to palliative banding of the pulmonary artery for heart failure early in life in patients with transposition of the great arteries and ventricular septal defect. Debanding and corrective operation later in the first year of life appear to offer a better chance for survival. It is also evident that the final chapter is not written, as these patients return with venous obstruction, some more than a year after operation. The position in this institution remains somewhat tentative for babies less than 4 months old, but the truly remarkable results now expected in babies 6 to 12 months of age provide sufficient encouragement to persist with early intraatrial baffle procedures by improved techniques in these patients. Conclusions in the miscellaneous group are difficult to draw because of the variety of lesions and small numbers of patients. It is clear, however, that in some patients with left-to-right shunts due to atrioventricular canal defects or doubleoutlet right ventricle, pulmonary vascular occlusive disease can progress rapidly and may not be interrupted by corrective operation, even early in life. Patients with atrial septal defect requiring operation in the first 2 months of life are likely to have a complicated course, and the results are unpredictable. The importance of accurate diagnosis of complex and unusual defects along with precise operation in these tiny patients cannot be overemphasized.
References 1 . Barbosa, R., Somerville,J., and Ross, D. Aorta-right pulmonary anastomosis: Long term problems after total correction (abstract). Am J Cardiol 33:125, 1974. 2. Barratt-Boyes, B. G., and Neutze, J. M. Primary repair of tetralogy of Fallot in infancy using profound hypothermia with circulatory arrest and limited cardiopulmonary bypass: A comparison with conventional two stage management. Ann Surg 178:406, 1973. 3. Bonchek, L. I., Anderson, R. P., Wood, J. A., Chapman, R. D., and Starr, A. Intracardiac surgery with extracorporeal circulation in infants. Ann Thorac Surg 17:280, 1974. 4. Breckenridge, I. M., Oelert, H., Graham, G. R., Stark, J., Waterston, D. J., and Bonham-Carter, R. E. Open-heart surgery in the first year of life.J Thorac Cardiovasc Surg 65:58, 1973. 5. Brunberg, J. A., Reilly, E. L., and Doty, D. B. Central nervous system consequences in infants of cardiac surgery using deep hypothermia and circulatory arrest. Circulation 49, 50 (Suppl II):60, 1974. 6. Carpena, C., and Subramanian, S. Management of the ventricular septal defect in transposition (abstract). Am J Cardiol 33: 130, 1974. 7. Ching, E., DuShane, J. W., McGoon, D.C., and Danielson, G. K. Total correction of cardiac anomalies in infancy using extracorporeal circulation.J Thmac Cardiovasc Surg 62:117, 1971. 8. Doty, D. B., and Ehrenhaft, J. L. Correspondence. Ann Thorac Surg 17:210, 1974. 9. Kilman, J. W., Williams, T. E., Kakos, G. S., Craenen, J., and Hosier, D. M. Surgical correction of the transposition complex in infancy. J Thorac Cardiovasc Surg 66:387, 1973. 10. Mori, A., Muraoka, R., Yokata, Y., Okamoto, Y., Ando, F., Fukumasu, H., Oku, H., Ikeda, M., Shirotani, H., and Hikasa, Y. Deep hypothermia combined with cardiopulmonary bypass for cardiac surgery in neonates and infants. J Thorac Cardiovasc Surg 64:422, 1972. 1 1. Nadas, A. S., Fyler, D. C., and Castaneda, A. R. The critically ill infant with congenital heart disease. Mod C o w Cardiovasc Dis 42:53, 1973. 12. Pacifico, A. D., Bargeron, L. M., and Kirklin, J. W. Primary total correction of tetralogy of Fallot in children less than four years of age. Circulation 48:1085, 1973.
324
T H E ANNALS OF THORACIC SURGERY
Congenital Cardiac Anmalies in Infancy 13. Pierce, W. S., Raphaely, R. C., Downes, J. J., and Waldhausen, J. A. Cardiopulmonary bypass in infants: Indications, methods, and results in 32 patients. Surgery 70:839, 1971. 14. Rittenhouse, E. A., Mohri, H., Dillard, D. H., and Merendino, K. A. Deep hypothermia in cardiovascular surgery. Ann Thorac Surg 17:63, 1974. 15. Stark, J., Hucin, B., Aberdeen, E., and Waterston, D. Cardiac surgery in the first year of life: Experience with 1,049 operations. Surgery 69:483, 1971. 16. Stark, J., de Leval, M. R., Waterston, D. J., Graham, G. R., and Bonham-Carter, R. E. Corrective surgery of transposition of the great arteries in the first year of life.J Thorac Cardiovasc Surg 67:673, 1974. 17. Starr, A., Bonchek, L. I., and Sunderland, C. 0. Total correction of tetralogy of Fallot in infancy. J Thorac Cardiovasc Surg 65:45, 1973. 18. Stewart, S., Edmunds, L. H., Kirklin, J. W., and Allarde, R. R. Spontaneous breathing with continuous positive airway pressure after open intracardiac operations in infants. J Thorac Cardiovasc Surg 65:37, 1973. 19. Subramanian, S., and Wagner, H. Correction of transposition of the great arteries in infants under surface-induced deep hypothermia. Ann Thorac Surg 16:391, 1973. 20. Truccone, N. J., Bowman, F. O., Malm, J. R., and Gersony, W. M. Systemicpulmonary arterial shunts in the first year of life. Circulation 49:508, 1974. 21. Venugopal, P., Olszowka, J., Wagner, H., Vlad, P., Lambert, E., and Subramanian, S. Early correction of congenital heart disease with surface-induced deep hypothermia and circulatory arrest. J Thwac Cardiovasc Surg 66:375, 1973. 22. Venugopal, P., and Subramanian, S. Early primary repair of tetralogy of Fallot (abstract). Am J Cardiol 33:451, 1974. 23. Waterston, D. J., Stark, J., and Ashcraft, K. W. Ascending aorta-to-right pulmonary artery shunts: Experience with 100 patients. Surgery 72:897, 1972.
VOL. 20, NO. 3, SEPTEMBER, 1975
325
C
orrection of congenital cardiac anomalies has traditionally been performed in children when they reach the age of 5 or 6 years. By this time the child is generally of sufficient size that the technical aspects of operation are minimized and the deleterious effects of cardiopulmonary bypass can be distributed over a large body mass. If a child required operative intervention earlier in life, generally some form of palliative operation, such as banding of the pulmonary artery or systemic-to-pulmonary artery shunt, was performed because of the high mortality rates attending primary repair of these defects in infants and young children [4, 71. The results of palliative operation [ 15, 20, 231 and the problems attending the required second-stage operative repair [ 11 made this approach less than optimal. The advent of better blood oxygenators and blood filtering devices as well as the improved ability to manage cardiopulmonary bypass, often combined with hypothermia and circulatory arrest techniques, have allowed primary, one-stage repairs of a variety of congenital cardiac defects in infants. A number of highly successful series were reported in the period 1971-1973 [lo, 11, 13, 18,211,andthesubjecthasbeenreviewedbyRittenhouse [14] and Bonchek [3] and their associates. From the Divisions of Thoracic and Cardiovascular Surgery, Department of Surgery, and of Pediatric Cardiology, Department of Pediatrics, T h e University of Iowa Hospitals and Clinics, Iowa City, Iowa. Accepted for publication Feb. 27, 1975. Address reprint requests to Dr. Doty, Division of Thoracic and Cardiovascular Surgery, Department of Surgery, The University of Iowa Hospitals and Clinics, Iowa City, Iowa. 52242. 316
T H E ANNALS OF THORACIC SURGERY
Congenital Cardiac Anomalies in Infancy Encouraged by the results of others, a program of intracardiac repair of congenital cardiac anomalies, regardless of age at the time of necessary operative intervention, was elected at The University of Iowa Hospitals. All patients in whom repair was feasible were accepted, and none was excluded because of associated anomalies or general condition. This report details the results in 74 consecutive patients in order to determine if early repair of congenital cardiac defects improves results.
Clinical Information and Methods There were 74 patients with a variety of congenital cardiac anomalies requiring operative intervention during the period February 1, 1972, to September 15, 1974 (see Table 1).The age of the patients ranged from 14 hours to 24 months, and they weighed 2.4 to 14.7 kg (Figure). Growth was retarded in most of the patients, with 61 falling below the sixteenth percentile for weight. All these patients had intracardiac repair of their defect. The only palliative operations performed during this period were done in patients in whom correction was not feasible. The indications for operation in patients with defects having predominantly left-to-right intracardiac shunts were intractable congestive cardiac failure or advancing pulmonary vascular resistance. Patients with anomalies producing arterial desaturation were operated upon when hypoxic spells occurred or if cyanosis worsened and polycythemia advanced. There were 20 patients with a hematocrit over 60% and 11 over 70%; the highest was 82%. Two techniques were used to manage the circulation during operation. The first group of 28 consecutive patients were operated on using deep hypothermia and circulatory arrest for the intracardiac repair with a limited period of extracorporeal circulation for cooling and rewarming. This standardized technique was utilized in order to collect data on the central nervous system consequences and has been reported elsewhere [5].After a low incidence of permanent nervous system damage was documented in this initial group, hypothermia and arrest techniques continued to be utilized with confidence in selected cases, especially in very small patients (less than 6 kg) and in those with complex intraatrial defects. A total of 51 patients had their circulation arrested to facilitate operative repair. There were, in addition, 23 patients in whom the circulation was continuously maintained by cardiopulmonary bypass. These were patients in whom the intracardiac repair could be expected to require less than 40 minutes, the exposure to repair the defect would be straightforward, and who were generally larger (weight greater than 6 kg). In these patients, circulatory arrest seemed to provide no great advantage over continuous bypass with moderate hypothermia; an initial 5-minute period with the perfusate at 15°C was followed by a period of aortic occlusion up to40 minutes, during which the perfusion temperature was 28°C and the flow to the patient 1.8 liters per minute per square meter of body surface area. Choice of perfusion technique or the use of circulatory arrest did not appear to affect mortality, as deaths were related to the cardiac anomaly and its repair rather than to the life-support technique used during the operation. Standard methods of intracardiac repair of defects were used in all patients VOL.
20, NO. 3, SEPTEMBER, 1975
317
DOTY, LAUER, AND EHRENHAm WEIGHT
Percentile 84 50 16
t
0
I
I
I
I
1 2 3 4
I
I
6 12 AGE (months)
I
18
1
24
Age and weight of 74 patients having early correction of congenital cardiac anomalies.Dots within circles repfesent early death, showing that most occurred inpatients less than 4 months old. Predicted weighfor age is superimposed by percentile to emphasize the marked growthfailure in these patients.
except for 14 with transposition of the great arteries who had the intraatrial baffle constructed of tubular weave-knit Dacron [8]. This has now been discarded in favor of pericardium. Elective operation was performed in 59 patients and emergent procedures in 15. The latter group is particularly noteworthy in that 12 subsequently died. Two of these patients had cardiorespiratory arrest prior to the procedure, and 1 underwent operation with a heart rate less than 40. Six of the patients had pulmonary edema and were on the respirator, one with a Pk0, of 114 mm Hg in spite of ventilatory assistance. Seven patients had a Pao, below 27 mm Hg, 1 of whom had seizures and a Pao, of 17 mm Hg. A second intracardiac operation was required in 5 patients, making a total of 79 procedures in these 74 patients. The follow-up period was 2 weeks to 30 months .
Results Since this is not a uniform group of patients, the results are analyzed according to the lesion. Table 1 presents the mortality occurring during the 30 days after operation. There were 15 deaths, which is 20% of the patients. Ten of the 15 early deaths occurred in babies less than 4 months old (see the Figure). The cause of death in 8 patients who died later is shown in Table 2. VENTRICULAR SEF'TAL DEFECT
Twelve patients had closure of ventricular septal defect because of intractable heart failure often associated with growth retardation. There were no hospital or late deaths. The age range was 17 days to 24 months with 5 patients less than 4 months old. Two patients had associated anomalous muscle bands in the right ventricle that obstructed the outflow and required excision, and 3 had multiple muscular septal defects. Nearly all the patients had high pulmonary blood flow, and pulmonary resistance ranged from 1.36 to 6.0 units. The ratio of pulmonary to systemic resistance was greater than 0.25 in 6 patients, the highest being 0.66. Two patients have small residual shunts(QJQ 1.16 to 1.3). One child has bron318
THE ANNALS OF THORACIC SURGERY
Congenital Cardiac Anomalies in Infancy TABLE 1. HOSPITAL MORTALITY FOLLOWING REPAIR OF CONGENITAL CARDIAC ANOMALIES IN PATIENTS 24 MONTHS OF AGE OR LESS
Anomaly Ventricular septal defect Tetralogy of Fallot With pulmonary atresia Transposition of great arteries Simple Complex Miscellaneous Double-outlet right ventricle Atrial septal defect Coarctation, VSD, ASD, PDA Total anomalous pulmonary veins Pulmonary stenosis Aortic stenosis Atrioventricular canal (complete) Hemitruncus Cor triatriatum Aortopulmonary window, VSD, anomalous coronary artery Total
No. of Patients
Mortality
< 30 Days 0
12 16 2
2
14 11
4 4
3
0 0 1
2 2
1
3 2 2 2 1 1 1
1 0 1 0 0 0
74
15
1
VSD = ventricular septal defect; ASD = atrial septal defect; PDA = patent ductus arteriosus.
chial asthma which is limiting development, and another has continued symptoms of left ventricular dysfunction. The others have done well in terms of general physical condition and exercise tolerance; growth improvement has been unpredictable. TETRALOGY OF FALLOT
The 18 patients in this group represented the full spectrum of the anomaly. The age range was 4 to 24 months with only 4 patients less than 12 months old. The indications for operation were hypoxic spells in 12 patients and increasing cyanosis and polycythemia in 6. Patients were selected for repair of the defect based on angiographic anatomy that suggested the feasibility of satisfactory anatomical correction. Generally, this meant pulmonary arteries of apparently “good’size and the pulmonary valve annulus measuring at least one-third of the aortic valve annular diameter. Only 2 patients had right ventricular hypoplasia and were classified as having pulmonary atresia because there was barely continuity between the right ventricle and pulmonary artery through the markedly hypoplastic pulmonary annulus. Since both patients died following patch reconstruction of the right ventricular outflow tract in addition to the usual tetralogy of Fallot repair, no further patients with unfavorable anatomy were operated on for repair. During this period a total of 9 aortopulmonary shunt procedures have been performed at this institution for palliation of patients with severe forms of tetralogy of Fallot or pulmonary atresia. There were 2 deaths (mortality, 22%). Seven of the 9 patients were less than 8 weeks old, and 5 operations were done in VOL. 20, NO. 3, SEPTEMBER, 1975
3 19
DOTY, LAUER, AND EHRENHAET TABLE 2. CAUSES OF 8 LATE DEATHS
Anomaly Transposition of great arteries Simple Complex Double-outlet right ventricle Atrial septal defect Atrioventricular canal (complete) Cor triatriatum
Months after Operation
6 14 1% 5
7 4
3
2
Cause of Death
.
Pulmonary venous obstruction Pulmonary venous obstruction Heart failure Superior vena cava obstruction Obstructive pulmonary vascular disease Sudden death, cause unknown Obstructive pulmonary vascular disease & mitral valve incompetence Pulmonary venous obstruction
the first week of life. Shunt procedures were generally favored in patients less than 4 months of age, and the anatomy was critically analyzed in all patients less than 12 months old when considering the choice of procedure when operative intervention was required, Using these rigid criteria in the selected 16 patients with favorable anatomy operated upon for repair of tetralogy of Fallot, only 1 died (mortality, 6%).Except for a patient who developed choreoathetosis after operation, all are doing remarkably well. TRANSPOSITION OF T H E GREAT ARTERIES
There were 25 patients 14 hours to 22 months of age in the group with transposition of the great arteries. Fourteen of the patients were less than 6 months old. A Rashkind balloon atrial septostomy was performed at the initial cardiac catheterization in 2 1 patients, a Blalock-Hanlon atrial septectomy was done at another institution in 1, and 3 patients had large ventricular septal defects so that mixing of pulmonary and systemic blood was thought to be adequate at initial catheterization.. Five patients did not improve following Rashkind atrial septostomy and underwent repair within 48 hours after cardiac catheterization. Of those improved by septostomy, the salutary clinical effect lasted from 1 to 2 1 months with an average of 7 months until surgical intervention was required. Fourteen had transposition of the great arteries with intact ventricular septum and no associated anomaly other than patent ductus arteriosus in 6. Transposition of the great arteries with associated infundibular pulmonary stenosis was present in 4, ventricular septal defect in 4, and both pulmonary stenosis and ventricular septal defect in 3 patients. Eight of the patients died less than 48 hours after operation (32%). Five of these were treated in the initial 4 months' experience. Four patients died 1 ?4to 14 months later (16%).A number of factors appeared to be associated with mortality. Age was very important, in that 9 of the 12 deaths were in patients who underwent operation at less than 4 months. All 4 patients having the intraatrial baffle operation (Mustard procedure) when Rashkind septostomy failed during the first week of life died, 2 of them shortly after operation and 1 at 5 weeks due to low 320
THE ANNALS OF THORACIC SURGERY
Congenital Cardiac Anomalies in Infancy cardiac output with apparently technically sound procedures; the fourth died at 14 months from pulmonary venous obstruction. Technical aspects of the operation and placement of the intraatrial partition to redirect venous return were important factors in 8 of the 12 deaths. Two patients had unrelieved obstructing hypertrophy of the outflow areas in the right or left ventricle. In 2 patients the pericardial baffle was redundant and appeared to obstruct pulmonary venous outflow. Another patient died of supraventricular tachyarrhythmia, perhaps related to the placement of the suture line relative to the conduction pathways. Three late deaths were due to pulmonary venous or superior vena cava obstruction, probably related to the configuration of a Dacron intraatrial conduit. One patient died of central nervous system complications 2 days after operation. Seizures had been marked prior to operation, and the hematocrit was 82% with a PaOz of 17 mm Hg. The convulsive disorder had been thought to be hypoxic in origin, but postmortem examination revealed sagittal sinus and cortical vein thromboses that had been present prior to operation. Reoperation for pulmonary venous or superior vena cava obstruction was performed in 4 patients, all of whom had Dacron intraatrial conduits. Two subsequently died and 2 are presently well. A good result was achieved in 13 patients with follow-up to 23 months. There was a remarkable improvement in general well-being and exercise tolerance as well as a dramatic change in the children’s appearance. MISCELLANEOUS ANOMALIES
The miscellaneous group includes 19 patients with a variety of cardiac anomalies. The experience with these patients is too limited for conclusions to be drawn regarding individual anomalies, but the group considered as a whole allows some general impressions. Complex anomalies and young babies seemed the rule, with an age range of 2 days to 22 months and 9 of the 19 under 4 months old. Double-Outlet Right Ventricle. Two patients with subaortic ventricular septal defect have done well while a third with indeterminate ventricular septal defect died 7 months after operation. This patient had advanced pulmonary arteriolar occlusive disease, marked desaturation, and polycythemia with a hematocrit of 72%. The pulmonary resistance of 9.3 units remained abnormal after operation, and progressive dilation of the right ventricle, especially the outflow portion, preceded death. Atrial Septal Defect. One patient had cardiorespiratory arrest and congestive cardiac failure prior to closure of the atrial septal defect and died suddenly 4 months after operation; no cause was identified. Another patient, operated on because of cardiomegaly and heart failure, has continued to have cardiac enlargement and mitral valve incompetence and had a presumed cerebral embolism with right hemiparesis 2 months after operation. The nervous system defect subsequently cleared, but the clinical course suggests myocardiopathy. Ventricular and Atrial Sepal Defect, Patent Ductus Arterww, and Coarctation of the Aorta. This difficult combination of anomalies has been repaired in a single VOL. 20, NO. 3, SEPTEMBER, 1975
321
DOTY, LAUER, AND EHRENHAFT operation using both sternotomy and lateral thoracotomy incisions in 2 patients. Encouraged by the first patient, who had an excellent result, the operation was attempted in a 1-month-old infant with respiratory insufficiency due to left-toright shunt. In spite of respiratory therapy, the Pacon continued to rise and was 114 mm Hg prior to operation. The patient died 24 hours after the procedure with what appeared to be low cardiac output and a hemorrhagic disorder due to disseminated intravascular coagulation. Total Anomalous Pulmonary Venous Connection. One patient with a subdiaphragmatic connection and another with connection to the coronary sinus did well, and the result was dramatic. Another patient with the supracardiac variety died because a patent ductus arteriosus was present which was not ligated. Pulmonary Valve Stenosis. Two patients with this anomaly did well; 1 of them had associated defects of the atrial and ventricular septa and patent ductus arteriosus. Aortic Valve Stenosis. One patient, 2 days old, was operated upon after a cardiac arrest and in spite of adequate valvulotomy died in the operating room. Postmortem examination showed severe endocardia1 fibroelastosis of the left ventricle. Another patient, 2 months old, did well. Complete Atrioventricular Canal. Two patients were successfully operated upon at 13 and 22 monthsof age, respectively. However, 1 of them, who had heart failure, cyanosis, and pulmonary vascular resistance of 8.8 units, did not improve. Subsequent cardiac catheterization showed systolic right ventricular pressure to be 110 mm Hg and left ventricular pressure 96 mm Hg. There was a right-to-left shunt (Q&& 0.65) and mitral valve incompetence. The child died 3 months postoperatively. Anomalous Origin of Right Pulmonary Arteryfrom Aorta. This anomaly (hemitruncus arteriosus) was successfully corrected in 1 patient by disconnecting the right pulmonary artery from the aorta and creating an anastomosis to the main pulmonary artery. Cor Triatriatum. Successful operation at 5 weeks in 1 patient with initial improvement was followed by progressive pulmonary venous obstruction. The patient died 2 months later at a second operation. Scarring of the pulmonary venous orifices and moderate pulmonary venous hypoplasia were found. Aortofmlmonary Window, Ventricular Septa1 Defect, and Anomalous Oripn of Coronary Arteriesfrom Pulmonary Arteries. The coronary artery anomaly was not detected prior to operation in 1 patient but was diagnosed and properly repaired by patch closure of the aortopulmonary window in such fashion as to bring the coronary arteries in continuity with the aorta through the window. Myocardial ischemia time, however, resulted in tetanic contracture of the ventricles that could not be reversed.
Comment The hypothesis that one-stage repair of congenital cardiac anomalies will be associated with improved results when operative intervention is required, regard322
T H E ANNALS OF THORACIC SURGERY
Congenital Cardiac Anomalies in Infancy less of age, has been tested. N o attempt was made to favor the results in this group of patients, as all were accepted regardless of general condition if complete repair was feasible. Perhaps the results would have been better if those with cardiac arrest, respiratory failure, or central nervous system damage had been eliminated from the study. By accepting all patients into the series, a more representative sample of the benefits of this approach was tested. Subgroup analysis allows some conclusions regarding the utility of primary correction of congenital heart defects in infancy. In the ventricular septal defect group, it is clear that early operation is safe and is preferable to palliative banding of the pulmonary artery. The uniform survival and good late results in our patients agree completely with the experience in a large group reported by Mori and associates [ 101. Even in patients with complex and multiple ventricular septal defects who undergo operative intervention early in life, good results can be anticipated. Similarly, tetralogy of Fallot can be repaired early with operative mortality rates as satisfactory as with the traditional two-stage approach. Certainly, postoperative respiratory management seems easier in patients who have had less time to develop extensive aortopulmonary collateral circulation. This experience is supported by the results of others [12, 17, 221, so early one-stage repair may be performed in these patients provided that the anatomy, especially the size of the pulmonary annulus, permits it. It should be noted, however, that judging the adequacy of the pulmonary annulus may be difficult in patients less than 4months old. This small experience does not justify extension of early repair to patients with severe hypoplasia of the right ventricular outflow tract or pulmonary atresia, as suggested hy Barratt-Boyes and Neutze 121. The overall results in our patients with transposition were not as favorable as in the experience of others [9, 191. A number of factors are probably responsible for the increased mortality, not the least of which was the technical conduct of the operation. Technical problems were associated with 8 of the 12 deaths, 5 of which occurred at the beginning of the series when understanding of these matters was less precise. Since we adopted Subramanian’s modification [19] of the Mustard operation, the results have clearly improved. This technique, using pericardium, takes a final configuration remarkably similar to the Brom modification, which Stark and colleagues [16] have used so successfully in small patients. The age of patients having the Mustard procedure is very important to the results achieved. As shown in this series, 8 of the 12 deaths occurred in patients less than 4 months old. Increased mortality has also been observed by Subramanian and Wagner [ 191 in patients with simple transposition of the great arteries, in patients less than 1 month old, and in patients less than 6 months old with associated complex anomalies. It should probably be predicted therefore that this series so heavily weighted with young babies (over half less than 6 months old) would have an increased mortality. The complex factors responsible for failure of the Rashkind procedure to achieve palliation in the first week of life apparently were still present following the intraatrial baffle operation, contributing to results so poor that this approach has been abandoned. Similar experience has directed Carpena VOL. 20, NO. 3, SEPTEMBER, 1975
323
DOTY, LAUER, AND EHRENHAFT and Subramanian [6] back to palliative banding of the pulmonary artery for heart failure early in life in patients with transposition of the great arteries and ventricular septal defect. Debanding and corrective operation later in the first year of life appear to offer a better chance for survival. It is also evident that the final chapter is not written, as these patients return with venous obstruction, some more than a year after operation. The position in this institution remains somewhat tentative for babies less than 4 months old, but the truly remarkable results now expected in babies 6 to 12 months of age provide sufficient encouragement to persist with early intraatrial baffle procedures by improved techniques in these patients. Conclusions in the miscellaneous group are difficult to draw because of the variety of lesions and small numbers of patients. It is clear, however, that in some patients with left-to-right shunts due to atrioventricular canal defects or doubleoutlet right ventricle, pulmonary vascular occlusive disease can progress rapidly and may not be interrupted by corrective operation, even early in life. Patients with atrial septal defect requiring operation in the first 2 months of life are likely to have a complicated course, and the results are unpredictable. The importance of accurate diagnosis of complex and unusual defects along with precise operation in these tiny patients cannot be overemphasized.
References 1 . Barbosa, R., Somerville,J., and Ross, D. Aorta-right pulmonary anastomosis: Long term problems after total correction (abstract). Am J Cardiol 33:125, 1974. 2. Barratt-Boyes, B. G., and Neutze, J. M. Primary repair of tetralogy of Fallot in infancy using profound hypothermia with circulatory arrest and limited cardiopulmonary bypass: A comparison with conventional two stage management. Ann Surg 178:406, 1973. 3. Bonchek, L. I., Anderson, R. P., Wood, J. A., Chapman, R. D., and Starr, A. Intracardiac surgery with extracorporeal circulation in infants. Ann Thorac Surg 17:280, 1974. 4. Breckenridge, I. M., Oelert, H., Graham, G. R., Stark, J., Waterston, D. J., and Bonham-Carter, R. E. Open-heart surgery in the first year of life.J Thorac Cardiovasc Surg 65:58, 1973. 5. Brunberg, J. A., Reilly, E. L., and Doty, D. B. Central nervous system consequences in infants of cardiac surgery using deep hypothermia and circulatory arrest. Circulation 49, 50 (Suppl II):60, 1974. 6. Carpena, C., and Subramanian, S. Management of the ventricular septal defect in transposition (abstract). Am J Cardiol 33: 130, 1974. 7. Ching, E., DuShane, J. W., McGoon, D.C., and Danielson, G. K. Total correction of cardiac anomalies in infancy using extracorporeal circulation.J Thmac Cardiovasc Surg 62:117, 1971. 8. Doty, D. B., and Ehrenhaft, J. L. Correspondence. Ann Thorac Surg 17:210, 1974. 9. Kilman, J. W., Williams, T. E., Kakos, G. S., Craenen, J., and Hosier, D. M. Surgical correction of the transposition complex in infancy. J Thorac Cardiovasc Surg 66:387, 1973. 10. Mori, A., Muraoka, R., Yokata, Y., Okamoto, Y., Ando, F., Fukumasu, H., Oku, H., Ikeda, M., Shirotani, H., and Hikasa, Y. Deep hypothermia combined with cardiopulmonary bypass for cardiac surgery in neonates and infants. J Thorac Cardiovasc Surg 64:422, 1972. 1 1. Nadas, A. S., Fyler, D. C., and Castaneda, A. R. The critically ill infant with congenital heart disease. Mod C o w Cardiovasc Dis 42:53, 1973. 12. Pacifico, A. D., Bargeron, L. M., and Kirklin, J. W. Primary total correction of tetralogy of Fallot in children less than four years of age. Circulation 48:1085, 1973.
324
T H E ANNALS OF THORACIC SURGERY
Congenital Cardiac Anmalies in Infancy 13. Pierce, W. S., Raphaely, R. C., Downes, J. J., and Waldhausen, J. A. Cardiopulmonary bypass in infants: Indications, methods, and results in 32 patients. Surgery 70:839, 1971. 14. Rittenhouse, E. A., Mohri, H., Dillard, D. H., and Merendino, K. A. Deep hypothermia in cardiovascular surgery. Ann Thorac Surg 17:63, 1974. 15. Stark, J., Hucin, B., Aberdeen, E., and Waterston, D. Cardiac surgery in the first year of life: Experience with 1,049 operations. Surgery 69:483, 1971. 16. Stark, J., de Leval, M. R., Waterston, D. J., Graham, G. R., and Bonham-Carter, R. E. Corrective surgery of transposition of the great arteries in the first year of life.J Thorac Cardiovasc Surg 67:673, 1974. 17. Starr, A., Bonchek, L. I., and Sunderland, C. 0. Total correction of tetralogy of Fallot in infancy. J Thorac Cardiovasc Surg 65:45, 1973. 18. Stewart, S., Edmunds, L. H., Kirklin, J. W., and Allarde, R. R. Spontaneous breathing with continuous positive airway pressure after open intracardiac operations in infants. J Thorac Cardiovasc Surg 65:37, 1973. 19. Subramanian, S., and Wagner, H. Correction of transposition of the great arteries in infants under surface-induced deep hypothermia. Ann Thorac Surg 16:391, 1973. 20. Truccone, N. J., Bowman, F. O., Malm, J. R., and Gersony, W. M. Systemicpulmonary arterial shunts in the first year of life. Circulation 49:508, 1974. 21. Venugopal, P., Olszowka, J., Wagner, H., Vlad, P., Lambert, E., and Subramanian, S. Early correction of congenital heart disease with surface-induced deep hypothermia and circulatory arrest. J Thwac Cardiovasc Surg 66:375, 1973. 22. Venugopal, P., and Subramanian, S. Early primary repair of tetralogy of Fallot (abstract). Am J Cardiol 33:451, 1974. 23. Waterston, D. J., Stark, J., and Ashcraft, K. W. Ascending aorta-to-right pulmonary artery shunts: Experience with 100 patients. Surgery 72:897, 1972.
VOL. 20, NO. 3, SEPTEMBER, 1975
325