- Email: [email protected]
Early Primary Repair of Tetralogy of Fallot
ORIGINAL ARTICLES
Early Primary Repair of Tetralogy of Fallot Robert A. Gustafson, M.D., Gordon F. Murray, M.D., Herbert E. Warden, M.D., Ronald C. Hill, M.D., and G. Edward Rozar, Jr., M.D. ABSTRACT Young age, low weight, and the requirement for transannular patch reconstruction of the right ventricular outflow tract (RVOT) are thought to adversely affect intracardiac repair of tetralogy of Fallot. Forty patients underwent complete repair between January, 1984, and January, 1987. Only infants with pulmonary atresia, complete atrioventricular canal, or coronary artery anomalies were shunted initially. The malalignment ventricular septal defect was closed with a Dacron patch. Infundibular resection was minimized. All atrial communications were left open. Thirty-four patients (85%) had a transannular RVOT patch, and 2 patients (5%) had a nontransannular patch. All 10 infants weighing 3.4 to 9.6 kg had a transannular RVOT patch at 7 weeks to 12 months of age. An RVOT patch was used in 26 of 30 children operated on between 1 and 6 years of age (median age, 24 months). No patient undergoing intracardiac repair died. Postoperative RVOT pullback gradients were between 0 and 35 mm Hg (mean, 18.5 mm Hg). Postoperative pulmonary artery saturation (mean, 73%) did not reveal any residual ventricular septal defect. The right ventriculadarterial pressure ratio was always less than 0.6 (mean, 0.4). All children are doing well at follow-up from two to 37 months. Serial echocardiograms reveal no residual ventricular septal defect and only 1 moderate RVOT gradient. Follow-up cardiac catheterization in 15 patients revealed no residual ventricular septal defect and RVOT gradients between 5 and 35 mm Hg. The right ventricular/left ventricular pressure ratio was always less than 0.6 (mean, 0.48). The early and late results justify continued primary repair of tetralogy of Fallot in symptomatic children, regardless of age or weight. A transannular RVOT patch did not have a major impact on operative mortality. Since Lillehei and colleagues [l] reported the intracardiac repair of tetralogy of Fallot in 1954 using controlled cross-circulation, the morbidity and mortality of repair have progressively declined with major advances in cardiac surgical techniques, postoperative intensive care, and cardiac anesthesia [2, 31. Many centers still advocate an initial aortopulmonary shunt for young patients with tetralogy of Fallot who need an operation From the Division of Cardiothoracic Surgery, Department of Surgery, West Virginia University Medical Center, Morgantown, WV. Presented at the Twenty-third Annual Meeting of the Society of Thoracic Surgeons, Toronto, Ont, Canada, Sept 21-23, 1987. Address reprint requests to Dr. Gustafson, Department of Surgery, West Virginia University Medical Center, Morgantown, WV 26505.
235
in the first few years of life, followed by intracardiac repair. Initial aortopulmonary shunts in infants are not without marked morbidity and mortality [4, 51. The selective use of primary repair in infants has been reported, but the criteria to determine optimal candidates for initial palliation versus complete repair in young children remain unclear [6, 71. Stimulated by the excellent results of Barratt-Boyes and Neutz [8] and Castaneda and colleagues [9] with primary repair in infancy, we adopted a policy in 1984 of early repair in all infants and children with symptomatic tetralogy of Fallot, regardless of their age or weight. Recently, transannular patch reconstruction, age, and weight have been implicated as independent risk factors for early primary repair [6, 101. In an effort to clarify the role of these risk factors, we reviewed all patients who had undergone repair of tetralogy in our center since 1984.
Patients and Methods From January, 1984, to January, 1987, at our center 40 consecutive patients underwent primary repair of symptomatic tetralogy of Fallot. Thirty boys and 10 girls ranged in age from 7 weeks to 6 years and weighed 3.8 to 24 kg (mean weight, 10.8 kg). Ten patients were less than 1 year of age. The median age of the other 30 patients was 24 months. Thirteen patients had associated cardiovascular lesions (Table 1).
lndications for Operation Early repair in infancy was performed for hypoxic spells in 5 patients and severe cyanosis with a hemoglobin concentration of 19 to 20 g d d l in 5 patients. In the 30 children over 1 year of age, the indication for primary repair was hypoxic spells in 5 patients and progressive cyanosis in 25 patients. Every attempt was made to electively repair all patients by age 2 years. Patients older than 2 years at initial presentation were referred for primary repair at that time. All patients underwent repair as indicated without regard to age, weight,. or branch pulmonary artery size. Contraindications to early repair in infancy were pulmonary atresia, complete atrioventricular canal defect, and anomalous origin of the left anterior descending coronary artery from the right coronary artery. Only these patients underwent an initial palliative shunt procedure. Patients with acyanotic tetralogy of Fallot with a net left-to-right shunt and an arterial saturation greater than 93% were excluded from this study. Surgical Technique Profound hypothermic circulatory arrest, with surface cooling and extracorporeal core cooling and rewarming,
A n n Thorac S u r g 45:235-241, M a r 1988. Copyright 0 1988 by T h e Society of Thoracic Surgeons
236 The Annals of Thoracic Surgery Vol45 No 3 March 1988
Table 1. Associated Cardiovascular Defects
Table 2 . Outflow Tract Reconstruction ~~
No. of
Defect Right aortic arch Bicuspid aortic valve
Subaortic stenosis Left superior vena cava Absent left pulmonary artery Secundum ASD ASD
=
Patients
Age (mo)
6 1 1
11-71
46
35
~~~
~
Age
No Patch
RV Patch
RV-PA Patch
<1 yr >1 yr
0
0
4 (13%)
2 (7%)
10 (100%) 24 (80%)
4 (10%)
2 (5%)
34 (85%)
Total
right ventricular; PA
pulmonary artery.
2-17 18
RV
1
2
17-32
type were primarily closed in 2 patients. The ductus arteriosus was ligated before cooling in patients who underwent hypothermic circulatory arrest. One patient also had resection of a discrete subaortic stenosis through the ascending aorta. Right atrial, left atrial, and main pulmonary artery catheters were inserted to improve postoperative monitoring and assist in better fluid management. These catheters were also used to determine intracardiac shunts and to measure right ventricular (RV)-pulmonary artery gradients in the immediate postoperative period. Temporary atrial and ventricular pacemaker electrodes were also placed.
3
atrial septal defect
was used for operations in infants and patients less than 10 kg (18 patients). Standard cardiopulmonary bypass with systemic hypothermia to 25°C was the technique used in older children. Myocardial protection was accomplished by cold potassium crystalloid cardioplegia and topical hypothermia. Myocardial temperatures were measured in the perfusion group, and additional cardioplegia was given as necessary to maintain a myocardial temperature less than 16°C. Hypothermic circulatory arrest was used in the 3 patients with a persistent left superior vena cava draining into the coronary sinus. The intracardiac repair was performed through a vertical incision in the right ventricular outflow tract (RVOT) parallel to the left anterior descending coronary artery. If the pulmonary valve annulus or main pulmonary artery was hypoplastic, the incision was carried across the pulmonary valve annulus to the bifurcation of the main pulmonary artery. The annular incision crossed the commissure between the anterior and right pulmonary valve cusp. If stenosis of the origin of the left pulmonary artery was present, the incision was continued onto the left pulmonary artery. A pulmonary valvotomy was performed through the main pulmonary artery if the pulmonary valve annulus was not hypoplastic. The infundibular incision was extended across the 0s infundibuli onto the trabecular portion of the right ventricle. Only large hypertrophied infundibular muscle bundles were resected. Infundibular resection was minimized in infants who required transannular patching. No attempt was made to "scoop out" the floor of the infundibulum. The large malalignment type of ventricular septal defect was closed with a Dacron patch using interrupted pledgeted sutures. In 34 patients the RVOT was closed with a transannular patch (Table 2). A nontransannular RVOT patch was used in 2 patients, and the right ventriculotomy was primarily closed in 4 patients. One patient who was found to have an unrecognized left anterior descending coronary artery arising from the right coronary artery underwent extensive infundibular resection and primary closure of the ventriculotomy. A patient with an absent left pulmonary artery had a transannular patch extending onto the right pulmonary artery. The atrial septum was usually inspected. A patent foramen ovale was not closed. Atrial septal defects of the secundum
=
=
FOllOW-up
Echocardiograms were obtained on all patients in the early postoperative period and just before discharge. Subsequently, serial echocardiogams were performed at six-month intervals. All patients were seen by an attending pediatric cardiologist at three-month intervals; a chest roentgenogram and an electrocardiogram were obtained on those visits. Cardiac catheterization was recommended at one year after repair.
Results
RVOT Reconstruction Of the 40 patients, 34 (85%) had hypoplasia of the pulmonary valve annulus or the main pulmonary artery, or both. A transannular pericardial patch was used to relieve the RVOT obstruction. All 10 infants weighing 3.8 to 8.6 kg had transannular patching. In 2 patients (5%), a nontransannular RV patch was sufficient to relieve the outflow obstruction. Infundibular resection and primary closure of the ventriculotomy were adequate in 4 patients (10%)(Table 2). All patients under 24 months of age required a transannular patch (Fig 1). A nontransannular patch or primary ventriculotomy closure was possible in 6 of the 20 patients (30%)older than 24 months. Figure 2 summarizes the incidence of outflow tract patching with regard to weight. All 19 patients weighing less than 10 kg had transannular patch reconstruction. Postoperative Complications All infants received digitalis and diuretics following repair. Continuous infusion of dopamine in the renal range of 2 to 4 pg/kg/min was frequently used in the first 24 hours after operation (60% of total group). Only 3 infants required ventilatory support beyond 48 hours.
237 Gustafson et al: Early Repair of Tetralogy of Fallot
15
0 Transannular
Nontransannular =No patch
-r 40 - 30 O
Highest
E E
Lowest
c 0
a 20
n
>
a 10
z
c
m
e
<1
1-2
2-4
0
)4
YEARS
Fig 1. Use of right ventricular outflow patches according to age of patients at time of operation. 15
Table 3. lrnrnediate (48 hours) Outflow Tract Gradient
0 Transannular
Nontransannular
UI
c
._ * n 0
0
2
5
C
Mild (15-39 mmHg)
Moderate (>40mm Hg)
3 (75%) 1(50%) 27 (79%)
1(25%) 1(50%)
4 2
7 (21%)
0 0 0
31 (78%)
9 (22%)
0
40
Trival
1c
4-6
6-9
10-15
k9
Fix 2 . Use of right ventricular outflow patches according to weight ofpatients at time of operation.
No patient sustained a neurological deficit. Serial electrocardiograms revealed a right bundle branch block pattern in 90% of the patients. No patient sustained surgical complete heart block.
Early Hernodynamic Results Hemodynamic and oximetric studies were obtained 48 hours after operation in all patients, using the intracardiac catheters placed at repair. Simultaneous right atrial and main pulmonary artery oxygen saturations were compared to identify any residual ventricular septal defects. The main pulmonary artery saturations ranged from 68 to 77% (mean, 73%). No marked hemodynamic shunt across the ventricular septum was identified. The immediate postoperative RVOT gradient was obtained 48 hours following operation. The right ventricle to radial arterial pressure ratio was also calculated. The early RVOT gradient ranged between 0 to 30 mm Hg (mean, 15 mm Hg; Fig 3). The transannular patch group varied between 0 to 25 mm Hg (mean, 12 mm Hg). The nontransannular patch group varied between 10 to 20 mm Hg (mean, 15 mm Hg), and the primary ventricular closure group ranged between 5 to 30 mm Hg (mean, 14 mm Hg). Table 3 summarizes the early hemodynamic results according to the degree of residual pulmonary outflow tract obstruction. The RV/arterial pressufe ratio ranged between 0.31 to 0.58 (mean, 0.41). There was no
Patch Placement
(G14 mmHg)
None RV patch RV-PA patch Total RV
=
right ventricular; PA
=
Total
34
pulmonary artery.
difference in the ratio according to the type of RVOT reconstruction. Early postoperative echocardiograms did not reveal any residual ventricular septal defects or marked residual pulmonary obstruction.
Late Hernodynamic Results Of the 40 patients, 15 (37%)have undergone postoperative cardiac catheterization between one and two years following repair. No residual ventricular septal defects were identified by oximetric studies or left ventriculography. No pulmonary artery aneurysms were found. The degree of pulmonary insufficiency was estimated by the technique reported by Oku and colleagues [ll].The 3 patients with no outflow tract patches had no pulmonary insufficiency. The 2 patients with nontransannular patches had only Grade 1 (mild) pulmonary insufficiency. The 10 patients who had transannular patch reconstruction had either Grade 1 or Grade 2 (moderate) pulmonary insufficiency. In the transannular group, the RV-pulmonary artery gradient ranged between 10 to 37 mm Hg (mean, 22 mm Hg; Fig 4). The range for the nontransannular patch group was 12 to 18 mm Hg (mean, 15 mm Hg), and the range for the group undergoing primary closure was 12 to 45 mm Hg (mean, 24 mm Hg). The patient with the 45 mm Hg gradient had the unrecognized left anterior descending coronary artery from the right coronary artery. The RV/left ventricular ratio was between 0.31 to 0.60 (mean, 0.45) in the transannular patch group, 0.32 to 0.36 (mean, 0.34) in
238 The Annals of Thoracic Surgery Vol 45 No 3 March 1988
-o 50 I
Highest
- 40 E
c
Lowest
Q
e a2 a
n
30
> a
I
20
5
10
3
I
9 0
Table 4. Late Outflow Tract Gradients at Recatheterization ~~
Patch Placement None RV patch RV-PA patch Total RV
=
Trival (0-14
Mild
~
Moderate (>40mm
mmHg)
(1539 mmHg)
2 (67%) 1(50%) 2 (20%)
0 1(50%) 8 (80%)
0 0
2 10
5 (33%)
9 (60%)
1(7%)
15
Hg)
1(33%)
Total
3
right ventricular; PA = pulmonary artery.
the nontransannular patch group, and 0.34 to 0.54 (mean, 0.42) in the no patch group. Table 4 summarizes the degree of postoperative pulmonary outflow stenosis in the recatheterization group.
Operative Results No patient died following primary repair of symptomatic tetralogy of Fallot. All patients remain asymptomatic over follow-up from 2 to 37 months. Serial electrocardiograms continue to show sinus rhythm in all patients. No patient is taking antiarrhythmic drugs. Serial chest roentgenograms do not reveal any pulmonary artery aneurysms. Serial echocardiograms continue to show no residual ventricular septal defects and only 1 patient with moderate RVOT obstruction (patient with left anterior descending coronary artery from right coronary artery).
Comment The optimal management of infants and children with symptomatic tetralogy of Fallot can be primary repair or initial palliative shunt procedure followed by secondary repair. Barratt-Boyes and Neutz [8] and Starr and colleagues [12] in 1973 began routine primary repair in young, asymptomatic infants and subsequently reported better results than those achieved with a twostage approach. Disturbed by the reported morbidity and mortality of palliative procedures in infancy [4],
Castaneda and associates [9] proceeded to perform primary repair of tetralogy in infants, regardless of age or weight. Their operative mortality on the first group of infants less than 1 year old was 7.0%, comparable to the operative risk of an initial aortopulmonary shunt procedure. Except for infants with a body surface area less than 0.35 m2 and an age less than 6 months, Kirklin and associates [lo] prefer primary repair so long as transannular patch reconstruction is not required. Proponents of a two-stage approach emphasize the low morbidity and mortality of an initial aortopulmonary shunt and the potential for growth of the pulmonary arteries, avoiding a subsequent transannular outflow tract reconstruction. While Kirklin and associates [lo] have shown that the presence of a shunt did not affect hospital mortality at secondary repair, Sade and co-workers [13] have shown that over time a systemic pulmonary artery shunt will not usually enlarge the hypoplastic conus or pulmonary valve annulus or both. Fourteen of their 24 patients who initially underwent a Waterston anastomosis needed a transannular outflow patch at the time of repair of tetralogy. Six more patients required a valved conduit because of acquired pulmonary atresia [13]. Likewise, Pacific0 and associates [14] reported a 71% incidence of patch enlargement of the pulmonary valve ring in patients who underwent repair after initial Waterston shunt. Early primary repair of tetralogy not only avoids the morbidity and mortality of aortopulmonary shunt procedures but over time may preserve optimal left ventricular function. Borow and co-workers [ 151 investigated left ventricular function in a group of satisfactorily repaired tetralogy patients using methoxamine to increase afterload. There was no significant difference in the resting cardiac index, left ventricular systolic pressure, left ventricular end-diastolic pressure, or left ventricular minute-work index among a group of patients who underwent repair before age 2 years and a group of patients who underwent repair after age 2 years. However, with afterload stress, older patients had abnormal work function curves with depressed slopes. These findings raised the possibility that early primary repair may well preserve postoperative left ventricular function. Van Praagh and colleagues [16] proposed that tetralogy of Fallot is really a monology. The basic anomaly appears to be underdevelopment of the subpulmonary infundibulum (conus) and its sequelae. Failure of the normal posterior inferior and rightward movement of the crista supraventricularis results in poor alignment between the abnormally located crista above, relative to the normally located ventricular septum, and the septal band below; hence, the ventricular septal defect in tetralogy is due to cristal malalignment and not to membranous septal deficiency [16]. Anderson and colleagues [171 believe the common feature of infundibular obstruction in tetralogy is due to anterior and cephalad deviation of the infundibular septum. The resulting obstruction may be exacerbated by hypertrophy of the infundibular septum, hypertrophy of the body of the
239
Gustafson et al: Early Repair of Tetralogy of Fallot
trabecula septomarginalis, or all of these. They go on to comment that relief of infundibular muscular obstruction alone is often insufficient to completely relieve the RVOT obstruction [17]. Conal septal hypoplasia and a malalignment ventricular septal defect were present in all 41 infants reported on by Castaneda and co-workers [9]. The area of greatest obstruction to pulmonary blood flow in the majority of their infants was at the pulmonary valve annulus. Their findings support the concept that marked RV hypertrophy with tetralogy of Fallot is secondary to outflow tract obstruction and is progressive with age. In a recent article, Hammon and co-workers [6] reported that diffuse hypoplasia of the pulmonary outflow tract or pulmonary atresia requiring a transannular patch reconstruction and age less than 1 year were both independent risk factors after repair of tetralogy. The operative mortality in their infants was 14.3% compared with 5.6% in older children. Transannular patch reconstruction in infants increased the mortality to 15.8% compared with 0% in older children without a transannular patch. Transannular patching was also an independent incremental risk factor in the Alabama experience reported by Kirklin and associates [lo]. In a review of 208 children who had repair of tetralogy of Fallot, there was no association between operative mortality and the use of a transannular patch in the Boston experience, as reviewed by Murphy and co-workers [18]. In their infant repair group, 70% of the patients required a transannular patch and 22% had a nontransannular patch placed. Transannular patches did not increase the operative mortality of early primary repair of tetralogy in infancy. Our experience would seem to support this observation that transannular patching does not increase operative mortality. There were no deaths in this series, which included 10 infants and a total of 34 patients with transannular patches. The results of early and late postoperative evaluations after early primary repair are very encouraging. No patient was found to have a residual ventricular septal defect by early oximetric studies or serial echocardiograms. A low incidence of hemodynamically important shunts across the ventricular septum was also noted by Castaneda and associates [9] after early repair. The hallmark of tetralogy in young children is hypoplasia of the conus and pulmonary valve annulus. Secondary hypertrophy of the septal and parietal bands occurs over time. Therefore, extensive resection of the parietal and septal bands in early repairs may weaken the function of the right ventricle or predispose to residual ventricular septal defects. Severe right hypertrophy in older patients may make optimal visualization of the malalignment ventricular septal defect difficult. Follow-up cardiac catheterization in 15 patients continued to show no residual ventricular septal defects. The RVOT obstruction was repaired successfully in all patients. There was no difference in the early mean outflow tract gradient among the transannular patch, nontransannular patch, or no patch group. The early
mean RV/arterial pressure ratio did not differ among the methods of RVOT reconstruction. The ratio was always below 0.6, with a mean of 0.41 for the total group. Although only 15 patients have had late hemodynamic studies, the pressure difference across the outflow tract and RV/left ventricular ratio has not increased with time. There remains no difference between the methods of managing the RVOT obstruction in tetralogy of Fallot. Late echocardiograms in all patients continue to support this observation. Murphy and co-workers “1, reporting on the late hemodynamic results after repair of tetralogy, found that 90% of infants repaired under hypothermic arrest had less than a 40 mm Hg outflow gradient and that there was no association between the residual gradient or the RV systolic pressure and type of RV outflow reconstruction. All patients have been carefully followed up with serial chest roentgenograms. The development of aneurysm of the outflow patch has not occurred over our short follow-up period. We, as others, believe that aneurysms develop in patients with residual obstruction to pulmonary blood flow [9]. If by using a transannular patch, the residual outflow tract gradient can be optimally reduced, the risk of late aneurysm should be minimized. Opponents of transannular patching point to the long-term risk of pulmonary insufficiency. The creation of pulmonary insufficiency is inherent in all methods of relief of outflow tract obstruction in tetralogy, even pulmonary valvotomy alone. Ellison and colleagues [19] found in experimental animals that isolated pulmonary insufficiency is well tolerated over years. Calder and associates [20] found pulmonary insufficiency was well tolerated clinically and hemodynamically, unless there was distal pulmonary stenosis or branch stenosis. Using mean right atrial pressure as an index of RV diastolic compliance and, therefore, the hemodynamic effect of pulmonary insufficiency, Murphy and associates [ 181 found no significant difference among different methods of RVOT reconstruction after repair of tetralogy. Oku and associates [ll]graded the degree of postoperative pulmonary insufficiency after repair of tetralogy from Grade 1 (trace) to Grade 4 (severe). Late postoperative hemodynamics and ventricular function were excellent in patients with a mild residual pulmonary stenosis and pulmonary insufficiency of Grade 2 or less, and it was poor in those with moderate residual pulmonary stenosis and marked pulmonary insufficiency. To prevent marked pulmonary insufficiency, they felt that the crosssectional area index of the pulmonary annulus should be 1.75 to 2.5 cm/m2 after a transannular patch was used [Ill. In a review of the postoperative pulmonary angiograms from our 15 recatheterized patients, the nontransannular and transannular patch group had either Grade 1 or Grade 2 pulmonary insufficiency. We concur that mild pulmonary insufficiency is preferable to residual outflow tract stenosis [21]. The transannular patch should be made only large enough to adequately relieve
240 The Annals of Thoracic Surgery Vol45 No 3 March 1988
the outflow tract obstruction. The excellent postoperative results of several series of patients undergoing repair of tetralogy support the benign course of isolated pulmonary insufficiency over several years [ 18, 211. The hemodynamic sequelae of pulmonary insufficiency over a lifetime remain unknown. Our data support the concept of early primary repair of symptomatic tetralogy of Fallot regardless of age or weight. The frequent utilization of transannular patching was not a risk factor in this small series. Every effort must be made to completely relieve the RVOT obstruction regardless of the need for an outflow tract patch. Longer follow-up will be necessary to evaluate the late consequences of mild to moderate pulmonary insufficiency and the true incidence of patch aneurysms.
References 1. Lillehei CW, Cohen M, Warden HE, et al: Direct vision intracardiac surgical correction of the tetralogy of Fallot, pentology of Fallot, and pulmonary atresia defects. Ann Surg 142:148, 1955 2. Ruzyllo W, Nihill MR, Mullins CE, McNamara DG: Hemodynamic evaluation of 221 patients after intracardiac repair of tetralogy of Fallot. Am J Cardiol 34:565, 1974 3. Pacifico AD, Bargeron LM, Kirklin JW: Primary total correction of tetralogy of Fallot in children less than four years of age. Circulation 48:1085, 1973 4. Greenwood RD, Nadas AS, Rosenthal A, et al: Ascending aorto-pulmonary artery anastomosis for cyanotic congenital heart disease. Am Heart J 94:14, 1977 5. Barbosa R, Somerville J, Ross D: Aorto-right pulmonary artery anastomosis: long-term problems and results after total correction. Am J Cardiol33:125, 1974 6. Hammon JW, Henry CL, Merrill WH, et al: Tetralogy of Fallot: Selective surgical management can minimize operative mortality. Ann Thorac Surg 40:280, 1985 7. Arciniegas E, Farooki ZQ, Hakimi M, et al: Early and late results of total correction of tetralogy of Fallot. J Thorac Cardiovasc Surg 80:770, 1980 8. Barratt-Boyes BG, Neutz JM: 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 9. Castaneda AR, Freed MD, Williams RG, Norwood WI: Repair of tetralogy of Fallot in infancy: early and late results. J Thorac Cardiovasc Surg 74:372, 1977 10. Kirklin JW, Blackstone EW, Pacifico AD, et al: Routine primary repair vs. two-stage repair of tetralogy of Fallot. Circulation 60:373, 1979 11. Oku H, Shirotani H, Sunakawa A, Yokoyama T: Postoperative long-term results in total correction of tetralogy of Fallot: hemodynamics and cardiac function. Ann Thorac Surg 41:413, 1986 12. Starr A, Bonchek LI, Sunderland CO: Total correction of tetralogy of Fallot in infancy. J Thorac Cardiovasc Surg 65:45, 1973 13. Sade R, Sloss J, Traves S, et al: Repair of tetralogy of Fallot after aortopulmonary anastomosis. Ann Thorac Surg 23:32, 1977 14. Pacifico AD, Kirklin JW, Blackstone EH: Surgical management of pulmonary stenosis in tetralogy of Fallot. J Thorac Cardiovasc Surg 74:382, 1977
15. Borow KM, Green LH, Castaneda AR, Keane JF: Left ventricular function after repair of tetralogy of Fallot and its relationship to age at surgery. Circulation 61:1150, 1980 16. Van Praagh RV, Van Praagh S, Nebesar RA, et al: Tetralogy of Fallot: underdevelopment of the pulmonary infundibulum and its sequelae. Am J Cardiol 26:25, 1970 17. Anderson RH, Allwork SP, Ho SY, et al: Surgical anatomy of tetralogy of Fallot. J Thorac Cardiovasc Surg 81:887, 1981 18. Murphy ID, Freed MD, Keane JF, et al: Hemodynamic results after intracardiac repair of tetralogy of Fallot by deep hypothermia and cardiopulmonary bypass. Circulation 62:Suppl 1:168, 1980 19. Ellison RG, Brown WJ Jr, Yeh TJ, Hamilton WF: Surgical significance of acute and chronic pulmonary valvular insufficiency. J Thorac Cardiovasc Surg 60:549, 1970 20. Calder AL, Barratt-Boyes BG, Brandt PWT, Neutze JM: Postoperative evaluation of patients with tetralogy of Fallot repaired in infancy. J Thorac Cardiovasc Surg 77:704, 1979 21. Finnegan P, Maider R, Patel RG, et al: Results of total correction of the tetralogy of Fallot. Br Heart J 38:934, 1976
Discussion w. HAMMON (Nashville, TN): I congratulate Dr. Gustafson and his colleagues on their excellent series of patients with tetralogy of Fallot treated with primary correction. Their 0% operative mortality is a goal to which all of us should aspire. Although these admirable results are open to great discussion, I would like to examine some of the factors that may have contributed to the low mortality and to critically examine the series in this light. First, this is a small series, with less than 50 patients, which does not diminish the 0% operating mortality but softens the impact. Most modern series of corrective operations for tetralogy of Fallot have operative mortalities well below 10% and good long-term results. I would hope Dr. Gustafson would continue to collect data and report his results again in several years. I note that 18 of the 40 patients were operated on using circulatory arrest as the perfusion technique. Our group would agree with this choice. It appears that myocardial protection is better when one uses circulatory arrest rather than conventional cardiopulmonary bypass, especially in cyanotic children. We have measured myocardial ATP levels in 10 patients operated on in the first year of life using circulatory arrest and compared them to another 10 patients operated on using cardiopulmonary bypass. ATP levels were better preserved in the circulatory arrest group than in those with standard cardiopulmonary bypass. We believe our findings may be related to the fact that myocardial rewarming is much less a problem with circulatory arrest, especially in cyanotic lesions like tetralogy of Fallot, than in conventional cardiopulmonary bypass. I would be interested in the authors’ views on the subject of myocardial protection related to perfusion technique in this series. Finally, the authors state in both the abstract and the manuscript that the results in this series justify the routine application of primary correction of tetralogy regardless of age. In my opinion, the fact that only 10 patients in this series were below 1 year of age and the exact number below 6 months of age was not specified does not justify this assertion. In several larger series of tetralogy repairs, including our own, younger age at primary repair, especially below 6 months, can be identified as an incremental risk factor for operative mortality. In neonates it appears that cardiopulmonary bypass in the DR. JOHN
241 Gustafson et al: Early Repair of Tetralogy of Fallot
presence of immature organ systems may be an additional risk factor. When one adds this factor to the known fact that infants who present with symptomatic tetralogy tend to have a more diffuse hypoplasia of the right ventricular outflow tract and will thus require more extensive reconstruction, operations in the first 3 to 6 months of life have to be considered in a different light. If no other therapy were available, this aggressive approach would be justified. However, the two-stage attack on the problem with preliminary shunting in infancy and elective repair before 2 years of age is a very reasonable plan of therapy. We have performed 43 Blalock-Taussig shunts in infants and have had 1 non-cardiac death since 1975 in our institution. Secondary repair has been performed in the vast majority of these patients with a combined mortality of 4.3%,similar to the mortality of 3% in a larger group of 200 patients with primary repair. We feel that this spectrum of tetralogy is quite varied and does not permit a blanket surgical approach, but instead a careful evaluation of each patient with the surgical therapy tailored to the extent of the pathology.
Again, our congratulations to the authors and our appreciation to the Society for the privilege of the floor. DR. GUSTAFSON: Certainly we realize that the series is not large enough to draw blanket conclusions, as Dr. Hammon has mentioned. We feel, however, that the results that we have gotten with early correction are as good as one can get with an initial Blalock-Taussig shunt. Certainly 10 infants is not a large enough series to validate that conclusion. We felt constrained by the fact that we had submitted the abstract and so we did not update our results. However, 7 additional patients under the age of 1 year have undergone correction since the time of the abstract and still we have no deaths in the series. We certainly agree with Dr. Hammon that hypothermic arrest is the way to go. If possible, we would prefer to operate on all patients with tetralogy of Fallot within the first year of life. Our mean age of operation, which was 2 years of age in the abstract, is now 16 months of age for all elective operations. Once again I would like to thank the Society for the opportunity to present these data.
Notice from the Southern Thoracic Surgical Association The thirty-fifth Annual Meeting of the Southern Thoracic Surgical Association will be held at the Marco Island Resort, Marco Island, FL, November 10-12, 1988. There will be a $125 registration fee for nonmember physicians except for guest speakers, authors and coauthors on the program, and residents. There will be a $50 registration fee for attendees of the Postgraduate Course on Saturday, November 12, 1988. The Postgraduate Course of the Southern Thoracic Surgical Association has been expanded to a full day and will provide in-depth coverage of thoracic surgical topics selected primarily as a means to enhance and broaden the knowledge of practicing thoracic and cardiac surgeons. Members wishing to participate in the Scientific Program should submit an original abstract and one copy by May 15, 1988, to Robert M. Sade, M.D., Program Chairman, Southern Thoracic Surgical Association, 111 East Wacker Dr, Chicago, IL 60601. Abstracts must be submitted on the Southern Thoracic Surgical Association
abstract submission form. These forms may be obtained from the Association’s office or in this issue of The Annals of Thoracic Surgery. All slides used during the presentation must be 35 mm. Manuscripts of accepted papers must be submitted to The Annals of Thoracic Surgery prior to the 1988 meeting or to the Secretary-Treasurer at the opening of the Scientific Session. Applications for membershp should be completed by July 1, 1988, and forwarded to the Southern Thoracic Surgical Association, 111 East Wacker Dr, Chicago, IL 60601.
Gordon F. Murray, M . D . Secretay-Treasurer Southern Thoracic Surgical Association Basic Science Center West Virginia University Medical Center Morgantown, W V 26506
Early Primary Repair of Tetralogy of Fallot Robert A. Gustafson, M.D., Gordon F. Murray, M.D., Herbert E. Warden, M.D., Ronald C. Hill, M.D., and G. Edward Rozar, Jr., M.D. ABSTRACT Young age, low weight, and the requirement for transannular patch reconstruction of the right ventricular outflow tract (RVOT) are thought to adversely affect intracardiac repair of tetralogy of Fallot. Forty patients underwent complete repair between January, 1984, and January, 1987. Only infants with pulmonary atresia, complete atrioventricular canal, or coronary artery anomalies were shunted initially. The malalignment ventricular septal defect was closed with a Dacron patch. Infundibular resection was minimized. All atrial communications were left open. Thirty-four patients (85%) had a transannular RVOT patch, and 2 patients (5%) had a nontransannular patch. All 10 infants weighing 3.4 to 9.6 kg had a transannular RVOT patch at 7 weeks to 12 months of age. An RVOT patch was used in 26 of 30 children operated on between 1 and 6 years of age (median age, 24 months). No patient undergoing intracardiac repair died. Postoperative RVOT pullback gradients were between 0 and 35 mm Hg (mean, 18.5 mm Hg). Postoperative pulmonary artery saturation (mean, 73%) did not reveal any residual ventricular septal defect. The right ventriculadarterial pressure ratio was always less than 0.6 (mean, 0.4). All children are doing well at follow-up from two to 37 months. Serial echocardiograms reveal no residual ventricular septal defect and only 1 moderate RVOT gradient. Follow-up cardiac catheterization in 15 patients revealed no residual ventricular septal defect and RVOT gradients between 5 and 35 mm Hg. The right ventricular/left ventricular pressure ratio was always less than 0.6 (mean, 0.48). The early and late results justify continued primary repair of tetralogy of Fallot in symptomatic children, regardless of age or weight. A transannular RVOT patch did not have a major impact on operative mortality. Since Lillehei and colleagues [l] reported the intracardiac repair of tetralogy of Fallot in 1954 using controlled cross-circulation, the morbidity and mortality of repair have progressively declined with major advances in cardiac surgical techniques, postoperative intensive care, and cardiac anesthesia [2, 31. Many centers still advocate an initial aortopulmonary shunt for young patients with tetralogy of Fallot who need an operation From the Division of Cardiothoracic Surgery, Department of Surgery, West Virginia University Medical Center, Morgantown, WV. Presented at the Twenty-third Annual Meeting of the Society of Thoracic Surgeons, Toronto, Ont, Canada, Sept 21-23, 1987. Address reprint requests to Dr. Gustafson, Department of Surgery, West Virginia University Medical Center, Morgantown, WV 26505.
235
in the first few years of life, followed by intracardiac repair. Initial aortopulmonary shunts in infants are not without marked morbidity and mortality [4, 51. The selective use of primary repair in infants has been reported, but the criteria to determine optimal candidates for initial palliation versus complete repair in young children remain unclear [6, 71. Stimulated by the excellent results of Barratt-Boyes and Neutz [8] and Castaneda and colleagues [9] with primary repair in infancy, we adopted a policy in 1984 of early repair in all infants and children with symptomatic tetralogy of Fallot, regardless of their age or weight. Recently, transannular patch reconstruction, age, and weight have been implicated as independent risk factors for early primary repair [6, 101. In an effort to clarify the role of these risk factors, we reviewed all patients who had undergone repair of tetralogy in our center since 1984.
Patients and Methods From January, 1984, to January, 1987, at our center 40 consecutive patients underwent primary repair of symptomatic tetralogy of Fallot. Thirty boys and 10 girls ranged in age from 7 weeks to 6 years and weighed 3.8 to 24 kg (mean weight, 10.8 kg). Ten patients were less than 1 year of age. The median age of the other 30 patients was 24 months. Thirteen patients had associated cardiovascular lesions (Table 1).
lndications for Operation Early repair in infancy was performed for hypoxic spells in 5 patients and severe cyanosis with a hemoglobin concentration of 19 to 20 g d d l in 5 patients. In the 30 children over 1 year of age, the indication for primary repair was hypoxic spells in 5 patients and progressive cyanosis in 25 patients. Every attempt was made to electively repair all patients by age 2 years. Patients older than 2 years at initial presentation were referred for primary repair at that time. All patients underwent repair as indicated without regard to age, weight,. or branch pulmonary artery size. Contraindications to early repair in infancy were pulmonary atresia, complete atrioventricular canal defect, and anomalous origin of the left anterior descending coronary artery from the right coronary artery. Only these patients underwent an initial palliative shunt procedure. Patients with acyanotic tetralogy of Fallot with a net left-to-right shunt and an arterial saturation greater than 93% were excluded from this study. Surgical Technique Profound hypothermic circulatory arrest, with surface cooling and extracorporeal core cooling and rewarming,
A n n Thorac S u r g 45:235-241, M a r 1988. Copyright 0 1988 by T h e Society of Thoracic Surgeons
236 The Annals of Thoracic Surgery Vol45 No 3 March 1988
Table 1. Associated Cardiovascular Defects
Table 2 . Outflow Tract Reconstruction ~~
No. of
Defect Right aortic arch Bicuspid aortic valve
Subaortic stenosis Left superior vena cava Absent left pulmonary artery Secundum ASD ASD
=
Patients
Age (mo)
6 1 1
11-71
46
35
~~~
~
Age
No Patch
RV Patch
RV-PA Patch
<1 yr >1 yr
0
0
4 (13%)
2 (7%)
10 (100%) 24 (80%)
4 (10%)
2 (5%)
34 (85%)
Total
right ventricular; PA
pulmonary artery.
2-17 18
RV
1
2
17-32
type were primarily closed in 2 patients. The ductus arteriosus was ligated before cooling in patients who underwent hypothermic circulatory arrest. One patient also had resection of a discrete subaortic stenosis through the ascending aorta. Right atrial, left atrial, and main pulmonary artery catheters were inserted to improve postoperative monitoring and assist in better fluid management. These catheters were also used to determine intracardiac shunts and to measure right ventricular (RV)-pulmonary artery gradients in the immediate postoperative period. Temporary atrial and ventricular pacemaker electrodes were also placed.
3
atrial septal defect
was used for operations in infants and patients less than 10 kg (18 patients). Standard cardiopulmonary bypass with systemic hypothermia to 25°C was the technique used in older children. Myocardial protection was accomplished by cold potassium crystalloid cardioplegia and topical hypothermia. Myocardial temperatures were measured in the perfusion group, and additional cardioplegia was given as necessary to maintain a myocardial temperature less than 16°C. Hypothermic circulatory arrest was used in the 3 patients with a persistent left superior vena cava draining into the coronary sinus. The intracardiac repair was performed through a vertical incision in the right ventricular outflow tract (RVOT) parallel to the left anterior descending coronary artery. If the pulmonary valve annulus or main pulmonary artery was hypoplastic, the incision was carried across the pulmonary valve annulus to the bifurcation of the main pulmonary artery. The annular incision crossed the commissure between the anterior and right pulmonary valve cusp. If stenosis of the origin of the left pulmonary artery was present, the incision was continued onto the left pulmonary artery. A pulmonary valvotomy was performed through the main pulmonary artery if the pulmonary valve annulus was not hypoplastic. The infundibular incision was extended across the 0s infundibuli onto the trabecular portion of the right ventricle. Only large hypertrophied infundibular muscle bundles were resected. Infundibular resection was minimized in infants who required transannular patching. No attempt was made to "scoop out" the floor of the infundibulum. The large malalignment type of ventricular septal defect was closed with a Dacron patch using interrupted pledgeted sutures. In 34 patients the RVOT was closed with a transannular patch (Table 2). A nontransannular RVOT patch was used in 2 patients, and the right ventriculotomy was primarily closed in 4 patients. One patient who was found to have an unrecognized left anterior descending coronary artery arising from the right coronary artery underwent extensive infundibular resection and primary closure of the ventriculotomy. A patient with an absent left pulmonary artery had a transannular patch extending onto the right pulmonary artery. The atrial septum was usually inspected. A patent foramen ovale was not closed. Atrial septal defects of the secundum
=
=
FOllOW-up
Echocardiograms were obtained on all patients in the early postoperative period and just before discharge. Subsequently, serial echocardiogams were performed at six-month intervals. All patients were seen by an attending pediatric cardiologist at three-month intervals; a chest roentgenogram and an electrocardiogram were obtained on those visits. Cardiac catheterization was recommended at one year after repair.
Results
RVOT Reconstruction Of the 40 patients, 34 (85%) had hypoplasia of the pulmonary valve annulus or the main pulmonary artery, or both. A transannular pericardial patch was used to relieve the RVOT obstruction. All 10 infants weighing 3.8 to 8.6 kg had transannular patching. In 2 patients (5%), a nontransannular RV patch was sufficient to relieve the outflow obstruction. Infundibular resection and primary closure of the ventriculotomy were adequate in 4 patients (10%)(Table 2). All patients under 24 months of age required a transannular patch (Fig 1). A nontransannular patch or primary ventriculotomy closure was possible in 6 of the 20 patients (30%)older than 24 months. Figure 2 summarizes the incidence of outflow tract patching with regard to weight. All 19 patients weighing less than 10 kg had transannular patch reconstruction. Postoperative Complications All infants received digitalis and diuretics following repair. Continuous infusion of dopamine in the renal range of 2 to 4 pg/kg/min was frequently used in the first 24 hours after operation (60% of total group). Only 3 infants required ventilatory support beyond 48 hours.
237 Gustafson et al: Early Repair of Tetralogy of Fallot
15
0 Transannular
Nontransannular =No patch
-r 40 - 30 O
Highest
E E
Lowest
c 0
a 20
n
>
a 10
z
c
m
e
<1
1-2
2-4
0
)4
YEARS
Fig 1. Use of right ventricular outflow patches according to age of patients at time of operation. 15
Table 3. lrnrnediate (48 hours) Outflow Tract Gradient
0 Transannular
Nontransannular
UI
c
._ * n 0
0
2
5
C
Mild (15-39 mmHg)
Moderate (>40mm Hg)
3 (75%) 1(50%) 27 (79%)
1(25%) 1(50%)
4 2
7 (21%)
0 0 0
31 (78%)
9 (22%)
0
40
Trival
1c
4-6
6-9
10-15
k9
Fix 2 . Use of right ventricular outflow patches according to weight ofpatients at time of operation.
No patient sustained a neurological deficit. Serial electrocardiograms revealed a right bundle branch block pattern in 90% of the patients. No patient sustained surgical complete heart block.
Early Hernodynamic Results Hemodynamic and oximetric studies were obtained 48 hours after operation in all patients, using the intracardiac catheters placed at repair. Simultaneous right atrial and main pulmonary artery oxygen saturations were compared to identify any residual ventricular septal defects. The main pulmonary artery saturations ranged from 68 to 77% (mean, 73%). No marked hemodynamic shunt across the ventricular septum was identified. The immediate postoperative RVOT gradient was obtained 48 hours following operation. The right ventricle to radial arterial pressure ratio was also calculated. The early RVOT gradient ranged between 0 to 30 mm Hg (mean, 15 mm Hg; Fig 3). The transannular patch group varied between 0 to 25 mm Hg (mean, 12 mm Hg). The nontransannular patch group varied between 10 to 20 mm Hg (mean, 15 mm Hg), and the primary ventricular closure group ranged between 5 to 30 mm Hg (mean, 14 mm Hg). Table 3 summarizes the early hemodynamic results according to the degree of residual pulmonary outflow tract obstruction. The RV/arterial pressufe ratio ranged between 0.31 to 0.58 (mean, 0.41). There was no
Patch Placement
(G14 mmHg)
None RV patch RV-PA patch Total RV
=
right ventricular; PA
=
Total
34
pulmonary artery.
difference in the ratio according to the type of RVOT reconstruction. Early postoperative echocardiograms did not reveal any residual ventricular septal defects or marked residual pulmonary obstruction.
Late Hernodynamic Results Of the 40 patients, 15 (37%)have undergone postoperative cardiac catheterization between one and two years following repair. No residual ventricular septal defects were identified by oximetric studies or left ventriculography. No pulmonary artery aneurysms were found. The degree of pulmonary insufficiency was estimated by the technique reported by Oku and colleagues [ll].The 3 patients with no outflow tract patches had no pulmonary insufficiency. The 2 patients with nontransannular patches had only Grade 1 (mild) pulmonary insufficiency. The 10 patients who had transannular patch reconstruction had either Grade 1 or Grade 2 (moderate) pulmonary insufficiency. In the transannular group, the RV-pulmonary artery gradient ranged between 10 to 37 mm Hg (mean, 22 mm Hg; Fig 4). The range for the nontransannular patch group was 12 to 18 mm Hg (mean, 15 mm Hg), and the range for the group undergoing primary closure was 12 to 45 mm Hg (mean, 24 mm Hg). The patient with the 45 mm Hg gradient had the unrecognized left anterior descending coronary artery from the right coronary artery. The RV/left ventricular ratio was between 0.31 to 0.60 (mean, 0.45) in the transannular patch group, 0.32 to 0.36 (mean, 0.34) in
238 The Annals of Thoracic Surgery Vol 45 No 3 March 1988
-o 50 I
Highest
- 40 E
c
Lowest
Q
e a2 a
n
30
> a
I
20
5
10
3
I
9 0
Table 4. Late Outflow Tract Gradients at Recatheterization ~~
Patch Placement None RV patch RV-PA patch Total RV
=
Trival (0-14
Mild
~
Moderate (>40mm
mmHg)
(1539 mmHg)
2 (67%) 1(50%) 2 (20%)
0 1(50%) 8 (80%)
0 0
2 10
5 (33%)
9 (60%)
1(7%)
15
Hg)
1(33%)
Total
3
right ventricular; PA = pulmonary artery.
the nontransannular patch group, and 0.34 to 0.54 (mean, 0.42) in the no patch group. Table 4 summarizes the degree of postoperative pulmonary outflow stenosis in the recatheterization group.
Operative Results No patient died following primary repair of symptomatic tetralogy of Fallot. All patients remain asymptomatic over follow-up from 2 to 37 months. Serial electrocardiograms continue to show sinus rhythm in all patients. No patient is taking antiarrhythmic drugs. Serial chest roentgenograms do not reveal any pulmonary artery aneurysms. Serial echocardiograms continue to show no residual ventricular septal defects and only 1 patient with moderate RVOT obstruction (patient with left anterior descending coronary artery from right coronary artery).
Comment The optimal management of infants and children with symptomatic tetralogy of Fallot can be primary repair or initial palliative shunt procedure followed by secondary repair. Barratt-Boyes and Neutz [8] and Starr and colleagues [12] in 1973 began routine primary repair in young, asymptomatic infants and subsequently reported better results than those achieved with a twostage approach. Disturbed by the reported morbidity and mortality of palliative procedures in infancy [4],
Castaneda and associates [9] proceeded to perform primary repair of tetralogy in infants, regardless of age or weight. Their operative mortality on the first group of infants less than 1 year old was 7.0%, comparable to the operative risk of an initial aortopulmonary shunt procedure. Except for infants with a body surface area less than 0.35 m2 and an age less than 6 months, Kirklin and associates [lo] prefer primary repair so long as transannular patch reconstruction is not required. Proponents of a two-stage approach emphasize the low morbidity and mortality of an initial aortopulmonary shunt and the potential for growth of the pulmonary arteries, avoiding a subsequent transannular outflow tract reconstruction. While Kirklin and associates [lo] have shown that the presence of a shunt did not affect hospital mortality at secondary repair, Sade and co-workers [13] have shown that over time a systemic pulmonary artery shunt will not usually enlarge the hypoplastic conus or pulmonary valve annulus or both. Fourteen of their 24 patients who initially underwent a Waterston anastomosis needed a transannular outflow patch at the time of repair of tetralogy. Six more patients required a valved conduit because of acquired pulmonary atresia [13]. Likewise, Pacific0 and associates [14] reported a 71% incidence of patch enlargement of the pulmonary valve ring in patients who underwent repair after initial Waterston shunt. Early primary repair of tetralogy not only avoids the morbidity and mortality of aortopulmonary shunt procedures but over time may preserve optimal left ventricular function. Borow and co-workers [ 151 investigated left ventricular function in a group of satisfactorily repaired tetralogy patients using methoxamine to increase afterload. There was no significant difference in the resting cardiac index, left ventricular systolic pressure, left ventricular end-diastolic pressure, or left ventricular minute-work index among a group of patients who underwent repair before age 2 years and a group of patients who underwent repair after age 2 years. However, with afterload stress, older patients had abnormal work function curves with depressed slopes. These findings raised the possibility that early primary repair may well preserve postoperative left ventricular function. Van Praagh and colleagues [16] proposed that tetralogy of Fallot is really a monology. The basic anomaly appears to be underdevelopment of the subpulmonary infundibulum (conus) and its sequelae. Failure of the normal posterior inferior and rightward movement of the crista supraventricularis results in poor alignment between the abnormally located crista above, relative to the normally located ventricular septum, and the septal band below; hence, the ventricular septal defect in tetralogy is due to cristal malalignment and not to membranous septal deficiency [16]. Anderson and colleagues [171 believe the common feature of infundibular obstruction in tetralogy is due to anterior and cephalad deviation of the infundibular septum. The resulting obstruction may be exacerbated by hypertrophy of the infundibular septum, hypertrophy of the body of the
239
Gustafson et al: Early Repair of Tetralogy of Fallot
trabecula septomarginalis, or all of these. They go on to comment that relief of infundibular muscular obstruction alone is often insufficient to completely relieve the RVOT obstruction [17]. Conal septal hypoplasia and a malalignment ventricular septal defect were present in all 41 infants reported on by Castaneda and co-workers [9]. The area of greatest obstruction to pulmonary blood flow in the majority of their infants was at the pulmonary valve annulus. Their findings support the concept that marked RV hypertrophy with tetralogy of Fallot is secondary to outflow tract obstruction and is progressive with age. In a recent article, Hammon and co-workers [6] reported that diffuse hypoplasia of the pulmonary outflow tract or pulmonary atresia requiring a transannular patch reconstruction and age less than 1 year were both independent risk factors after repair of tetralogy. The operative mortality in their infants was 14.3% compared with 5.6% in older children. Transannular patch reconstruction in infants increased the mortality to 15.8% compared with 0% in older children without a transannular patch. Transannular patching was also an independent incremental risk factor in the Alabama experience reported by Kirklin and associates [lo]. In a review of 208 children who had repair of tetralogy of Fallot, there was no association between operative mortality and the use of a transannular patch in the Boston experience, as reviewed by Murphy and co-workers [18]. In their infant repair group, 70% of the patients required a transannular patch and 22% had a nontransannular patch placed. Transannular patches did not increase the operative mortality of early primary repair of tetralogy in infancy. Our experience would seem to support this observation that transannular patching does not increase operative mortality. There were no deaths in this series, which included 10 infants and a total of 34 patients with transannular patches. The results of early and late postoperative evaluations after early primary repair are very encouraging. No patient was found to have a residual ventricular septal defect by early oximetric studies or serial echocardiograms. A low incidence of hemodynamically important shunts across the ventricular septum was also noted by Castaneda and associates [9] after early repair. The hallmark of tetralogy in young children is hypoplasia of the conus and pulmonary valve annulus. Secondary hypertrophy of the septal and parietal bands occurs over time. Therefore, extensive resection of the parietal and septal bands in early repairs may weaken the function of the right ventricle or predispose to residual ventricular septal defects. Severe right hypertrophy in older patients may make optimal visualization of the malalignment ventricular septal defect difficult. Follow-up cardiac catheterization in 15 patients continued to show no residual ventricular septal defects. The RVOT obstruction was repaired successfully in all patients. There was no difference in the early mean outflow tract gradient among the transannular patch, nontransannular patch, or no patch group. The early
mean RV/arterial pressure ratio did not differ among the methods of RVOT reconstruction. The ratio was always below 0.6, with a mean of 0.41 for the total group. Although only 15 patients have had late hemodynamic studies, the pressure difference across the outflow tract and RV/left ventricular ratio has not increased with time. There remains no difference between the methods of managing the RVOT obstruction in tetralogy of Fallot. Late echocardiograms in all patients continue to support this observation. Murphy and co-workers “1, reporting on the late hemodynamic results after repair of tetralogy, found that 90% of infants repaired under hypothermic arrest had less than a 40 mm Hg outflow gradient and that there was no association between the residual gradient or the RV systolic pressure and type of RV outflow reconstruction. All patients have been carefully followed up with serial chest roentgenograms. The development of aneurysm of the outflow patch has not occurred over our short follow-up period. We, as others, believe that aneurysms develop in patients with residual obstruction to pulmonary blood flow [9]. If by using a transannular patch, the residual outflow tract gradient can be optimally reduced, the risk of late aneurysm should be minimized. Opponents of transannular patching point to the long-term risk of pulmonary insufficiency. The creation of pulmonary insufficiency is inherent in all methods of relief of outflow tract obstruction in tetralogy, even pulmonary valvotomy alone. Ellison and colleagues [19] found in experimental animals that isolated pulmonary insufficiency is well tolerated over years. Calder and associates [20] found pulmonary insufficiency was well tolerated clinically and hemodynamically, unless there was distal pulmonary stenosis or branch stenosis. Using mean right atrial pressure as an index of RV diastolic compliance and, therefore, the hemodynamic effect of pulmonary insufficiency, Murphy and associates [ 181 found no significant difference among different methods of RVOT reconstruction after repair of tetralogy. Oku and associates [ll]graded the degree of postoperative pulmonary insufficiency after repair of tetralogy from Grade 1 (trace) to Grade 4 (severe). Late postoperative hemodynamics and ventricular function were excellent in patients with a mild residual pulmonary stenosis and pulmonary insufficiency of Grade 2 or less, and it was poor in those with moderate residual pulmonary stenosis and marked pulmonary insufficiency. To prevent marked pulmonary insufficiency, they felt that the crosssectional area index of the pulmonary annulus should be 1.75 to 2.5 cm/m2 after a transannular patch was used [Ill. In a review of the postoperative pulmonary angiograms from our 15 recatheterized patients, the nontransannular and transannular patch group had either Grade 1 or Grade 2 pulmonary insufficiency. We concur that mild pulmonary insufficiency is preferable to residual outflow tract stenosis [21]. The transannular patch should be made only large enough to adequately relieve
240 The Annals of Thoracic Surgery Vol45 No 3 March 1988
the outflow tract obstruction. The excellent postoperative results of several series of patients undergoing repair of tetralogy support the benign course of isolated pulmonary insufficiency over several years [ 18, 211. The hemodynamic sequelae of pulmonary insufficiency over a lifetime remain unknown. Our data support the concept of early primary repair of symptomatic tetralogy of Fallot regardless of age or weight. The frequent utilization of transannular patching was not a risk factor in this small series. Every effort must be made to completely relieve the RVOT obstruction regardless of the need for an outflow tract patch. Longer follow-up will be necessary to evaluate the late consequences of mild to moderate pulmonary insufficiency and the true incidence of patch aneurysms.
References 1. Lillehei CW, Cohen M, Warden HE, et al: Direct vision intracardiac surgical correction of the tetralogy of Fallot, pentology of Fallot, and pulmonary atresia defects. Ann Surg 142:148, 1955 2. Ruzyllo W, Nihill MR, Mullins CE, McNamara DG: Hemodynamic evaluation of 221 patients after intracardiac repair of tetralogy of Fallot. Am J Cardiol 34:565, 1974 3. Pacifico AD, Bargeron LM, Kirklin JW: Primary total correction of tetralogy of Fallot in children less than four years of age. Circulation 48:1085, 1973 4. Greenwood RD, Nadas AS, Rosenthal A, et al: Ascending aorto-pulmonary artery anastomosis for cyanotic congenital heart disease. Am Heart J 94:14, 1977 5. Barbosa R, Somerville J, Ross D: Aorto-right pulmonary artery anastomosis: long-term problems and results after total correction. Am J Cardiol33:125, 1974 6. Hammon JW, Henry CL, Merrill WH, et al: Tetralogy of Fallot: Selective surgical management can minimize operative mortality. Ann Thorac Surg 40:280, 1985 7. Arciniegas E, Farooki ZQ, Hakimi M, et al: Early and late results of total correction of tetralogy of Fallot. J Thorac Cardiovasc Surg 80:770, 1980 8. Barratt-Boyes BG, Neutz JM: 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 9. Castaneda AR, Freed MD, Williams RG, Norwood WI: Repair of tetralogy of Fallot in infancy: early and late results. J Thorac Cardiovasc Surg 74:372, 1977 10. Kirklin JW, Blackstone EW, Pacifico AD, et al: Routine primary repair vs. two-stage repair of tetralogy of Fallot. Circulation 60:373, 1979 11. Oku H, Shirotani H, Sunakawa A, Yokoyama T: Postoperative long-term results in total correction of tetralogy of Fallot: hemodynamics and cardiac function. Ann Thorac Surg 41:413, 1986 12. Starr A, Bonchek LI, Sunderland CO: Total correction of tetralogy of Fallot in infancy. J Thorac Cardiovasc Surg 65:45, 1973 13. Sade R, Sloss J, Traves S, et al: Repair of tetralogy of Fallot after aortopulmonary anastomosis. Ann Thorac Surg 23:32, 1977 14. Pacifico AD, Kirklin JW, Blackstone EH: Surgical management of pulmonary stenosis in tetralogy of Fallot. J Thorac Cardiovasc Surg 74:382, 1977
15. Borow KM, Green LH, Castaneda AR, Keane JF: Left ventricular function after repair of tetralogy of Fallot and its relationship to age at surgery. Circulation 61:1150, 1980 16. Van Praagh RV, Van Praagh S, Nebesar RA, et al: Tetralogy of Fallot: underdevelopment of the pulmonary infundibulum and its sequelae. Am J Cardiol 26:25, 1970 17. Anderson RH, Allwork SP, Ho SY, et al: Surgical anatomy of tetralogy of Fallot. J Thorac Cardiovasc Surg 81:887, 1981 18. Murphy ID, Freed MD, Keane JF, et al: Hemodynamic results after intracardiac repair of tetralogy of Fallot by deep hypothermia and cardiopulmonary bypass. Circulation 62:Suppl 1:168, 1980 19. Ellison RG, Brown WJ Jr, Yeh TJ, Hamilton WF: Surgical significance of acute and chronic pulmonary valvular insufficiency. J Thorac Cardiovasc Surg 60:549, 1970 20. Calder AL, Barratt-Boyes BG, Brandt PWT, Neutze JM: Postoperative evaluation of patients with tetralogy of Fallot repaired in infancy. J Thorac Cardiovasc Surg 77:704, 1979 21. Finnegan P, Maider R, Patel RG, et al: Results of total correction of the tetralogy of Fallot. Br Heart J 38:934, 1976
Discussion w. HAMMON (Nashville, TN): I congratulate Dr. Gustafson and his colleagues on their excellent series of patients with tetralogy of Fallot treated with primary correction. Their 0% operative mortality is a goal to which all of us should aspire. Although these admirable results are open to great discussion, I would like to examine some of the factors that may have contributed to the low mortality and to critically examine the series in this light. First, this is a small series, with less than 50 patients, which does not diminish the 0% operating mortality but softens the impact. Most modern series of corrective operations for tetralogy of Fallot have operative mortalities well below 10% and good long-term results. I would hope Dr. Gustafson would continue to collect data and report his results again in several years. I note that 18 of the 40 patients were operated on using circulatory arrest as the perfusion technique. Our group would agree with this choice. It appears that myocardial protection is better when one uses circulatory arrest rather than conventional cardiopulmonary bypass, especially in cyanotic children. We have measured myocardial ATP levels in 10 patients operated on in the first year of life using circulatory arrest and compared them to another 10 patients operated on using cardiopulmonary bypass. ATP levels were better preserved in the circulatory arrest group than in those with standard cardiopulmonary bypass. We believe our findings may be related to the fact that myocardial rewarming is much less a problem with circulatory arrest, especially in cyanotic lesions like tetralogy of Fallot, than in conventional cardiopulmonary bypass. I would be interested in the authors’ views on the subject of myocardial protection related to perfusion technique in this series. Finally, the authors state in both the abstract and the manuscript that the results in this series justify the routine application of primary correction of tetralogy regardless of age. In my opinion, the fact that only 10 patients in this series were below 1 year of age and the exact number below 6 months of age was not specified does not justify this assertion. In several larger series of tetralogy repairs, including our own, younger age at primary repair, especially below 6 months, can be identified as an incremental risk factor for operative mortality. In neonates it appears that cardiopulmonary bypass in the DR. JOHN
241 Gustafson et al: Early Repair of Tetralogy of Fallot
presence of immature organ systems may be an additional risk factor. When one adds this factor to the known fact that infants who present with symptomatic tetralogy tend to have a more diffuse hypoplasia of the right ventricular outflow tract and will thus require more extensive reconstruction, operations in the first 3 to 6 months of life have to be considered in a different light. If no other therapy were available, this aggressive approach would be justified. However, the two-stage attack on the problem with preliminary shunting in infancy and elective repair before 2 years of age is a very reasonable plan of therapy. We have performed 43 Blalock-Taussig shunts in infants and have had 1 non-cardiac death since 1975 in our institution. Secondary repair has been performed in the vast majority of these patients with a combined mortality of 4.3%,similar to the mortality of 3% in a larger group of 200 patients with primary repair. We feel that this spectrum of tetralogy is quite varied and does not permit a blanket surgical approach, but instead a careful evaluation of each patient with the surgical therapy tailored to the extent of the pathology.
Again, our congratulations to the authors and our appreciation to the Society for the privilege of the floor. DR. GUSTAFSON: Certainly we realize that the series is not large enough to draw blanket conclusions, as Dr. Hammon has mentioned. We feel, however, that the results that we have gotten with early correction are as good as one can get with an initial Blalock-Taussig shunt. Certainly 10 infants is not a large enough series to validate that conclusion. We felt constrained by the fact that we had submitted the abstract and so we did not update our results. However, 7 additional patients under the age of 1 year have undergone correction since the time of the abstract and still we have no deaths in the series. We certainly agree with Dr. Hammon that hypothermic arrest is the way to go. If possible, we would prefer to operate on all patients with tetralogy of Fallot within the first year of life. Our mean age of operation, which was 2 years of age in the abstract, is now 16 months of age for all elective operations. Once again I would like to thank the Society for the opportunity to present these data.
Notice from the Southern Thoracic Surgical Association The thirty-fifth Annual Meeting of the Southern Thoracic Surgical Association will be held at the Marco Island Resort, Marco Island, FL, November 10-12, 1988. There will be a $125 registration fee for nonmember physicians except for guest speakers, authors and coauthors on the program, and residents. There will be a $50 registration fee for attendees of the Postgraduate Course on Saturday, November 12, 1988. The Postgraduate Course of the Southern Thoracic Surgical Association has been expanded to a full day and will provide in-depth coverage of thoracic surgical topics selected primarily as a means to enhance and broaden the knowledge of practicing thoracic and cardiac surgeons. Members wishing to participate in the Scientific Program should submit an original abstract and one copy by May 15, 1988, to Robert M. Sade, M.D., Program Chairman, Southern Thoracic Surgical Association, 111 East Wacker Dr, Chicago, IL 60601. Abstracts must be submitted on the Southern Thoracic Surgical Association
abstract submission form. These forms may be obtained from the Association’s office or in this issue of The Annals of Thoracic Surgery. All slides used during the presentation must be 35 mm. Manuscripts of accepted papers must be submitted to The Annals of Thoracic Surgery prior to the 1988 meeting or to the Secretary-Treasurer at the opening of the Scientific Session. Applications for membershp should be completed by July 1, 1988, and forwarded to the Southern Thoracic Surgical Association, 111 East Wacker Dr, Chicago, IL 60601.
Gordon F. Murray, M . D . Secretay-Treasurer Southern Thoracic Surgical Association Basic Science Center West Virginia University Medical Center Morgantown, W V 26506