- Email: [email protected]
The management of severe subaortic stenosis, ventricular septal defect, and aortic arch obstruction in the neonate
The management of severe subaortic stenosis, ventricular septal defect, and aortic arch obstruction in the neonate Neonates with ventricular septal defect and aortic arch obstruction frequently have subaortic stenosis resulting from posterior deviation of the infundibular septum. Because the aortic anulus is often hypoplastic, making direct resection of the infundibular septum through the standard transaortic approach difficult, the optimal method of repair is uncertain. From September 1989 through November 1991, seven patients with ventricular septal defect, coarctation (n = 4~ or interrupted aortic arch (n = 3) and severe subaortic stenosis underwent repair with use of a technique that included transatrial resection of the infundibular septum. Their ages ranged from 5 to 63 days (median 15 days) and weights from 1.3 to 5.4 kg (mean 3.1 kg). Only one patient was older than 1 month. The systolic and diastolic ratios of the diameter of the left ventricular outflow tract to that of the descending aorta were 0.53 ± 0.09 mm (standard deviation) and 0.73 ± 0.11, respectively. At operation, the posteriorly displaced infundibular septum was partiaUy removed through a right atrial approach by resecting the superior margin of the ventricular septal defect up to the aortic anulus. The resulting enlarged ventricular septal defect was then closed with a patch to widen the subaortic area. In each patient the aortic arch was repaired by direct anastomosis. AU patients survived operation; there was one late death from noncardiac causes 3 months after repair. The survivors remain well from 3 to 14 months after repair (mean 8 months). All are in sinus rhythm and none has a residual ventricular septal defect One patient underwent successfnl balloon dilation of a residual aortic arch gradient late after repair. No patient has significant residual subaortic stenosis, although one has valvular aortic stenosis. This series suggests that in neonates with ventricular septal defect and severe subaortic stenosis resulting from posterior deviation of the infundibular septum, direct relief can be satisfactorily accomplished from a right atrial approach. This method provides effective widening of the left ventricular outflow tract and is superior to palliative techniques or conduit procedures. (J 'fHORAC CARDIOVASC SURG 1993;105:289-96)
Edward L. Bove, MD,a L. LuAnn Minich, MDb (by invitation), Ara K. Pridjian, MDa (by invitation), Flavian M. Lupinetti, MDa (by invitation), A. Rebecca Snider, MDb (by invitation), Macdonald Dick II, MDb (by invitation), and Robert H. Beekman III, MDb (by invitation), Ann Arbor, Mich.
h e management of the neonate with malalignment ventricular septal defect (VSD) and aortic arch obstrucFrom the Department of Surgery, Section of Thoracic Surgery; and the Department of Pediatrics, Division of Pediatric Cardiology," The University of Michigan School of Medicine, Ann Arbor, Mich. Read at the Seventy-second Annual Meeting of The American Association for Thoracic Surgery, Los Angeles, Calif., April 26-29, 1992. Address for reprints: Edward L. Bove, MD, 2120 Taubman Center, Box 0344,1500 East Medical Center Dr., Ann Arbor, MI 48109. Copyright © 1993 by Mosby-Year Book, Inc. 0022-5223/93/$1.00/+0.10
12/6/42461
tion is often complicated by the presence of subaortic obstruction resulting from posterior deviation of the infundibular or conal septum. 1 When this obstruction is severe, direct resection by the transaortic approach is difficult because of the small size of the ascending aorta and aortic anulus, which limits the surgical exposure. A number of palliative procedures have been advocated for these patients. They include the use of a left ventricle to descending aorta conduit and anastomosis of the proximal pulmonary artery to the ascending aorta to bypass the obstructed subaortic area.s 3 These procedures are limited by the need for extracardiac conduits and the necessi289
The Journal of Thoracic and Cardiovascular Surgery February 1993
2 9 0 Bove et al.
Table I. Clinical and echocardiographic findings
Patient
Age (days)
Weight (kg)
Aortic arch anatomy
Aortic Anulus size (mm)
Systole
Diastole
28 5 8 22 15 63 14
5.4 3.7 3.6 1.9 1.3 3.4 2.4
CoA CoA lAA,A CoA lAA,B CoA lAA,A
22 ± 20
3.1 ± 1.4
7.0 5.6 6.1 5.3 4.9 6.0 6.8 6.0 ± 0.76
0.65 0.42 0.55 0.49 0.50 0.65 0.46 0.53 ± 0.09
0.73 0.73 0.68 0.56 0.92 0.75 0.71 0.73 ± 0.11
1 2 3 4 5 6 7 Mean
± SD
Ratio LVOT/descending aorta
LVOT, Left ventricular outflow tract; GoA, coarctation; fAA, interrupted aortic arch; A, type A; B, type B; SD, standard deviation.
ty for reoperation. In this condition, however, there remains another surgical approach to the left ventricular outflow tract: through the right atrium and the VSD itself.4, 5 Exposure is generally satisfactory even in the small neonatal heart and may providea more optimal way of directly resecting the obstructing muscle. Because of the limitations imposed by transaortic resection and the use of extracardiac conduits, in September 1989, for patients undergoing VSD closure and primary reconstruction of aortic arch obstruction, we adopted the approach of transatrial resection of the left ventricular outflow tract in the presence of posterior deviation of the infundibular septum. The results in all patients undergoing this procedure at the University of Michigan Medical Center were reviewed and form the basis of this report. Patients and methods From September 1989 to November 1991, seven patients with VSD, aortic arch obstruction, and marked posterior deviation of the infundibular (conal) septum underwent complete repair at C. S. Mott Children's Hospital, the University of Michigan Medical Center, with the use of a technique 'that included transatrial resection of the infundibular septum (Table I). Their ages ranged from 5 to 63 days (median 15 days) and weights from 1.3 to 5.4 kg (mean 3.1 kg). Only one patient was older than 1 month of age. All patients underwent complete two-dimensional and Doppler echocardiography, and three patients also underwent cardiac catheterization as part of their preoperative evaluation (see Table I). In each patient the VSD was of the malalignment type with leftward displacement of the infundibular septum into the left ventricular outflow tract (Fig. I). Because of the presence ofthis nonrestrictive defect, preoperative Doppler measurements of left ventricular outflow velocities were normal. Since calculation of the peak instantaneous gradient from the Bernoulli equation does not reflect the severity of subaortic stenosis in this setting, the degree of obstruction was determined by using anatomic measurements from the two-dimensional echocardiogram. From the standard parasternal long-axis view, the left ventricular outflow dimension was measured from the most posterior
edge of the infundibular septum to the most anterior edge of the opposite left ventricular wall in both systole and diastole (Fig. 2). To provide an index for comparison, this dimension was divided by the descending aortic diameter measured in systole and diastole at the level of the diaphragm from the subcostal sagittal view. For all measurements, at least three cardiac cycles were analyzed and the results averaged. In a separate study from this institution, normal diastolic ratios of the left ventricular outflow tract to the descending aorta were determined to be 1.31 ± 0.14 (mean ± standard deviation) and ranged from 1.00 to 1.54 as measured in a group of 15 infants without heart disease," For the seven patients in this report, the diastolic left ventricular outflow tract to descending aortic ratios ranged from 0.56 to 0.92 (mean 0.73 ± 0.11) and were significantly lower than those in the normal group (p < 0.05). As an additional means of validating the severity of the subaortic stenosis in our patients, the systolic left ventricular outflow tract to descending aortic ratios were measured and compared with previously accepted echocardiographic and angiographic criteria. Severe subaortic obstruction has been considered to be present when that ratio was at or below 0.6. 7 By comparison, our patients ranged from 0.42 to 0.65 with a mean of 0.53 ± 0.09. Coarctation of the aorta was present in four patients and interrupted aortic arch in three patients (type A in two and type B in one). A nonrestrictive infundibular VSD was present in each patient. In two patients, slight extension into the inlet septum was noted, and in one the defect extended to the anulus of the pulmonary valve. Five patients had a patent ductus arteriosus. Two-dimensional and Doppler echocardiographic examinations were repeated in all patients after complete repair. The most recent studies were used for analysis and were performed from 7 days to 13 months after operation (mean 67 days). Because the resection was not circular, remnants of the infundibular septum were frequently seen extending into the subaortic area in the parasternal long-axis view while the left ventricular outflow tract appeared widely patent in other views. For this reason, comparisons of postoperative and preoperative ratios were not performed. Instead, the Doppler peak instantaneous gradient calculated from the Bernoulli equation (valid after VSD closure) was used to assess residual left ventricular outflow obstruction. The level of obstruction was determined by observing the location of the initial increase in velocity with use of color and pulsed Doppler techniques. Surgical technique. All procedures were performed using
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 2
Fig. 1. Preoperative left ventricular angiogram in lateral view demonstrates a large VSD with posterior displacement of the infundibular septum. There was severe coarctation of the aorta with hypoplasia of the transverse aortic arch. deep hypothermia and a single period of circulatory arrest ranging from 22 to 68 minutes (mean 50 minutes). A Y connector was placed in the arterial cannula to allow simultaneous perfusion of the ascending and descending aortas during coolingin those patients with interrupted aortic arch. A single dose of dilute blood cardioplegic solution (30 ml/kg) was administered after cardiopulmonary bypass was discontinued, and the arch vessels were occluded during circulatory arrest. In each patient the aortic arch obstruction was reconstructed by direct anastomosis with the use of continuous absorbable sutures. A right atriotomy was then made, and the VSD was exposed through the tricuspid valve(Fig. 3). A retraction suture was placed on the superior rim of the defect to facilitate exposure of the right coronary cusp of the aortic valve, and the infundibular septum was retracted into view. A wedge of septum was then resected until the anulus of the aortic valve was reached (Figs. 4 and 5). In one patient, resection of the infundibular septum resulted in a small laceration of the base of the right coronary cusp of the aortic valve. This was directly repaired with interrupted sutures through the same right atrial approach. The enlarged VSD was then closed with a suitably shaped patch from the right side of the septum (see Figs. 4 and 5). Cardiopulmonary bypass was reestablished, and systemic rewarming was begun. Bypass was discontinued at a rectal temperature of 35° to 37° C. A right ventriculotomy was not employedin any patient.
Results All patients survived operation and were discharged from the hospital. One late noncardiac death occurred in
Bove et al.
29 1
Fig. 2. Preoperative echocardiogram in long-axis view demonstrates the technique used for left ventricular outflow tract measurements. The left ventricular outflow dimension was measured from the most posterior edge of the infundibular septum to the most anterior edge of the opposite left ventricular wall in both systole and diastole. This dimension was divided by the descending aortic diameter measured in the same cardiac cycle at the level of the diaphragm from the subcostal sagittal view. For all measurements, at least three cardiac cycles were analyzed and the results averaged. LV, Left ventricle; RV, right ventricle; LA, left atrium; Ao, aorta. a premature infant with multiple noncardiac problems 3 months after operation. The six late survivors remain clinically well from 3 to 14 months after repair (mean 8 months). Five of the patients require no cardiac medications. All patients are in normal sinus rhythm, and none has evidence of left bundle branch block. Follow-up two-dimensional and Doppler echocardiographic examination results are shown in Table II. Total peak instantaneous left ventricular outflow tract gradients ranged from 0 to 85 mm Hg, with five of the patients demonstrating no residual gradient. The only patient with a significant gradient had no evidence of subaortic obstruction because the increase in outflow velocity was located at the level of the aortic valve. This patient underwent balloon dilatation 13 months after repair, at which time the catheter gradient was found to be 120 mm Hg. This was reduced to 70 mm Hg after dilatation. No patient had significant subaortic stenosis, although mild turbulence was frequently noted beneath the aortic valve in the area of the resection. The only patient in this series with any subaortic gradient (25 mm Hg) was also the smallest patient in the group (1.3 kg), and in this patient resection of the infundibular septum was the most difficult. Residual aortic arch Doppler gradients were present in two patients, while the remaining five patients had no residual obstruction. One patient underwent successful balloon dilation of the aortic anastomosis 5 months after repair. There were no significant residual ventricular sep-
The Journal of Thoracic and Cardiovascular Surgery February 1993
2 9 2 Rove et at.
Infundibular septum -----,_:;'+_
'---------=--------
Fig. 3. Artist's depictiondemonstratesthe surgeon'sviewof the VSD through the right atrium and tricuspidvalve. A traction suture placed in the infundibular septum facilitates the exposure of the aortic valveanulus (inset).
Fig. 4. A wedgeof infundibularseptum is resecteduntil the anulus of the aortic valveis reached (left). The resulting enlarged VSD is then closed with a patch (right). tal defects. Aortic regurgitation by Doppler echocardiographic examination was absent in four patients and trivial in the remaining three.
Discussion In 1971, Van Praagh and associates'' first described the mechanism of obstruction to left ventricular outflow
resulting from posterior displacement of the infundibular or conal septum in neonates with VSD and aortic arch obstruction. This condition results from malalignment between the infundibular septum with the anterior and posterior extensions of the septal band. The infundibular septum inserts to the left of the posterior limb of the septal band rather than between the posterior and anterior
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 2
limbs. Thus a large malalignment VSD results, and muscular obstruction to left ventricular outflow may be present. The pulmonary valve anulus often overrides the VSD into the left ventricle. This situation is the opposite of that seen in tetralogy of Fallot in which the infundibular septum is displaced anteriorly and to the right of the anterior limb of the septal band, resulting in right ventricular outflow tract obstruction. Van Praagh and others 4• 8 postulated that successful complete repair would be possible if the obstructing muscle were resected. As successful operative treatment of patients with VSD and interrupted aortic arch or coarctation became common, the management of clinically significant subaortic obstruction assumed greater importance. Earlier attempts at reconstruction of the aortic arch combined with pulmonary artery banding were abandoned in many centers because of the finding that the muscular left ventricular outflow tract obstruction rapidly progressed, a processaccelerated by the increased afterload imposed by the band.? Ilbawi and associates'" reported 10 patients with interrupted aortic arch and posterior displacement of the infundibular septum. The four patients in their series who underwent pulmonary artery banding and arch repair all died, whereas all six who underwent placement of a graft between the descending aorta and the pulmonary artery proximal to the band survived. When the displacement of the infundibular septum is severe, arch repair and VSD closure alone may also produce poor results because of unrelieved subaortic obstruction. In an attempt to avoid this situation, alternate surgical approaches have been utilized. Placement of a conduit from the apex of the left ventricle to the descending aorta has been performed in the neonate to bypass the left ventricular outflow tract obstruction.' Alternatively, Yasui and colleagues' reported two infants who underwent successful complete repair by insertion of an intraventricular baffle channeling the VSD to the pulmonary valve, combined with anastomosis of the transected end of the main pulmonary artery to the ascending aorta. The aortic arch was repaired with a graft, and a right ventricle to distal pulmonary artery conduit was inserted to complete the repair. All operative approaches that include insertion of conduits from the right or left ventricle have significant limitations imposed by the lack of potential growth and the need for reoperation. When the aortic valve anulus is adequate in size, direct resection of the posteriorly displaced infundibular septum would appear to be the ideal approach to relieve left ventricular outflow tract obstruction. However, the technical limitations imposed by the frequently small size of the ascending aorta and the aortic valve anulus in this condition make transaortic
Bove et ai. 2 9 3
Fig. 5. Appearance of the VSD and displaced infundibular septumas seenin the lateralview (seeFigs. I and 2 for comparison). Position of the traction suture used to expose the aortic valve during resection (upper left); position of the VSD patch (lower right).
resection of the obstructing muscle difficult. 10 The possibility of approaching the subaortic obstruction through the VSD was proposed by Freedom and associates." This technique was successfully applied by DeLeon and his colleagues- in three infants, two of whom had a transatrial approach to the VSD and subaortic area. The benefits of this technique are substantial and include preservation of the native left ventricular outflow tract as well as avoidance of prosthetic grafts and conduits. Prediction of which patients with VSD and posterior displacement of the infundibular septum will require a more aggressive operative approach to relieve potential obstruction to left ventricular outflow is not always accurate. Some patients with mild degrees of subaortic obstruction will be adequately treated with arch repair and VSD closure alone. Measurement of left ventricular to aortic gradients is not a reliable index of the severity of subaortic stenosis when there is a nonrestrictive VSD or patent ductus arteriosus. Thus there exists a need to define criteria that accurately predict those patients in whom significant early left ventricular outflow tract obstruction from preoperative angiographic or echocardiographic dimensions, or both, will develop. Previously accepted angiographic criteria have included a systolic ratio of the
The Journal of Thoracic and Cardiovascular Surgery February 1993
2 9 4 Bove et al.
Table II. Postoperative two-dimensional and Doppler echocardiographic follow-up Patient
Total LVOT gradient (mm Hg)
Aortic arch gradient (mm Hg)
85
0 0 0
I 2 3 4 5
25
55 25
6 7
0 0
0 0
0 0 0
AR
Residual VSD
0 0
0 0
Trivial Trivial
Tiny
0 Trivial
0
0 0
Comments Gradient at valve level Trivial AR preoperatively, gradient subvalvular Aortic valve injury, balloon dilation residual arch gradient Gradient subvalvular, late death from aspiration
Tiny Tiny
LVOT, Left ventricular outflow tract; AR, aortic regurgitation; VSD, ventricular septal defect.
diameter of the left ventricular outflow tract to that of the descending aorta at the level of the diaphragm of 0.6 or less." Iwahara and associates'! used standard echocardiographic parasternal long-axis views to assess the severity of obstruction in a group of patients with aortic arch obstruction and malalignment VSD and compared them with another group with arch obstruction but without a malalignment VSD. They calculated the systolic ratio of the dimension of the left ventricular outflow tract proximal to the aortic valve to that of the ascending aorta immediately distal to the aortic valve. Those with a malalignment VSD had a mean ratio of 0.59 ± 0.09, which was significantly smaller than that of those patients without a malalignment VSD, 1.03 ± 0.11. Furthermore, the operative mortality among those patients with a malalignment VSD was 56% compared with 0% among those without septal malalignment. These authors concluded that the left ventricular outflow tract to ascending aorta ratio was a useful indicator of the severity of the immediate postoperative left ventricular outflow tract obstruction and that it correlated with the operative outcome. Our method is similar to that of Iwahara, although we chose to use the descending aorta in the ratio. Although it may be argued that the descending aorta may not be reliable in the face of aortic arch obstruction, five of our seven patients had descending aortic flow maintained by a large ductus arteriosus. On the other hand, normalizing the left ventricular outflow dimension to the ascending aorta may be even more questionable since there is often hypoplasia of the ascending aorta in this condition. Additionally, we have found that systolic measurements are not as consistently obtained as those measured in diastole. Thus the diastolic dimensions were also measured for each patient and compared with a group of infants without heart dlsease." The normal infants were found to have significantly larger ratios, 1.31 ± 0.14 versus 0.73 ± 0.11 (p < 0.05). Further confirmation of the accuracy of our method in predicting those patients with significant
subaortic stenosis can be found by analyzing the results in three additional patients who underwent operation for malalignment VSD and aortic arch obstruction in our institution during the same time interval of this study but in whom no infundibular septal resection was done. One patient with mild posterior displacement and a diastolic left ventricular to descending aortic ratio of 1.0 had no left ventricular outflow gradient after operation. The other two patients with ratios of 0.73 and 0.89 had gradients of 36 and 30 mm Hg, respectively. Thus we are confidant that all of the patients in this series had severe posterior displacement of the infundibular septum and would be predicted to have a significantly higher operative mortalityor to demonstrate a greater degree of residual postoperative subaortic obstruction in the absence of any infundibular septal resection. This is further substantiated by the work of Goldmuntz, Baffa, and Weinberg,'? who recommended primary repair for patients with VSD and interrupted aortic arch only when the subaortic region was greater than or equal to 4 mm in the long-axis view. By comparison, the corresponding dimension in our patients was 2.9 ± 0.7 mm (standard deviation) and ranged from 2.1 mm to 3.9 mm. Although the technique of resection of the left ventricular outflow tract obstruction through the VSD in these patients is technically easier than the transaortic approach, there remains the potential for serious complications. Exposure of the aortic valve from the right atrium is difficult, particularly in the tiny neonate, and operative injury to the right coronary cusp of the aortic valve (as occurred in one patient in this series) is a definite hazard. The use of a traction suture to evert the septum into view facilitates the exposure and reduces the possibilityof injury. It must be emphasized that our follow-up is short, and there remains a possibility that subaortic obstruction will progress and require additional intervention in the future. A number of previously reported series have stressed the likelihood of this occurrence and the importance of con-
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 2
tinued follow-up.P'!" However, the virtual complete absence of significant residual subaortic gradients in our patients, all of whom had severe displacement of the infundibular septum, suggests that early recurrence of obstruction is unlikely. The results in this series demonstrate that resection of left ventricular outflow tract obstruction in the neonate with aortic arch obstruction and a malalignment VSD can be effectively accomplished from a transatrial approach. The ability to relieve the obstruction without the use of external conduits is a significant advantage in this condition. We recommend transatrial resection of the infundibular septum when the diastolic left ventricular outflow tract to descending aorta ratio is less than or equal to 1.0 or the systolic ratio is less than or equal to 0.65. Furthermore, the safety and efficacy of this technique appear to justify its application even in patients with milder forms of septal displacement in whom preoperative uncertainty may exist as to the severity of the left ventricular obstruction that may result after repair. REFERENCES I. Moulaert AJ, Bruins CC, Oppenheimer-Dekker A. Anomalies of the aortic arch and ventricular septal defects. Circulation 1976;53:1011-5. 2. Norwood WI, Lang P, Castaneda AR, Murphy JD. Management of infants with left ventricular outflowobstruction by conduit interposition between the ventricular apex and thoracic aorta. J THORAC CARDIOVASC SURG 1983; 86:771-6. 3. Yasui H, Kado H, Nakano E, et al. Primary repair of interrupted aortic arch and severe aortic stenosis in neonates. J THORAC CARDIOVASC SURG 1987;93:539-45. 4. Freedom RM, Bain HH, Esplagas E, Dische R, Rowe RD. Ventricular septal defect in interruption of aortic arch. Am J Cardiol 1977;39:572-82. 5. DeLeon SY, Ilbawi MN, Roberson DA, et al. Conal enlargement for diffuse subaortic stenosis. J THORAC CARDIOVASC SURG 1991;102:814-20. 6. Minich LL, Snider AR, Bove EL, Lupinetti FM. Echo predictors of the need for infundibular wedge resection in infants with aortic arch obstruction, ventricular septal defect and subaortic stenosis. J Am Coil Cardiol 1992; 19:148A. 7. Hossack KF, Neutze JM, Lowe JB, Barratt-Boyes BG. Congenital valvar aortic stenosis: natural history and assessment for operation. Br Heart J 1980;43:561-73. 8. Van Praagh R, Bernhard WF, Rosenthal A, Parisi LF, Fyler DC. Interrupted aortic arch: surgical treatment. Am J CardioI1971;27:200-11. 9. Matitiau A, Geva T, Colan SD, et al. Bulboventricular foramen size in infants with double-inlet left ventricle or tricuspidatresia with transposed great arteries: influenceon initial palliative operation and rate of growth. J Am Coil CardioI1992;19:142-8.
Hove et af.
295
10. Ilbawi MN, Idriss FS, DeLeon SY, Muster AJ, Benson DW Jr, Paul MH. Surgical management of patients with interrupted aortic arch and severe subaortic stenosis. Ann Thorac Surg 1988;45:174-80. II. Iwahara M, Ino T, Nishimoto K, et al. Clinical features of aortic arch anomaly with malalignment ventricular septal defect. Ann Thorac Surg 1989;48:693-6. 12. Goldmuntz E, Baffa JM, Weinberg PM. Evaluation of the subaortic region in interrupted aortic arch. J Am Coli Cardiol 1992;19:233A. 13. Jones M, Barnhart GR, Morrow AG. Late results after operations for left ventricular outflowtract obstruction. Am J Cardiol 1982;50:569-79. 14. Attie F, OvseyevityJ, Buendia A, et al. Surgical results in subaortic stenosis. Int J CardioI1986;11:329-35. 15. Scott WA, Rocchini AP, Bove EL, et al. Repair of interrupted aortic arch in infancy. J THORAC CARDIOVASC SURG 1988;96:564-8. 16. Sell JE, Jonas RA, Mayer JE, Blackstone EH, Kirklin JW, Castaneda AR. The results of a surgical program for interrupted aortic arch. J THORAC CARDIOVASC SURG 1988; 96:864-77.
Discussion Dr. Cristo I. Tchervenkov (Montreal, Quebec, Canada). I believethat in any malformations associated with hypoplasia of the left heart structures there is a continuum in the degree of severity. Beyond the feasibility of resection of the conal septum, the size of the aortic valve anulus may affect the short-term outcome of this procedure in terms of survival and possibly the long-term outcome in terms of recurrence of left ventricular outflow tract obstruction. In addition to that, what should be done for patients who have aortic valvular stenosis as well? At the Montreal Children's Hospital, we have in 1989 dealt with two patients who had severe left ventricular outflow tract obstruction associated with posterior malalignment VSD and interrupted aortic arch. The marked deviation of the conal septum resulted in a subaortic area of about 3 mm and an aortic valve anulus of 4.5 mm in diameter. In these patients we have performed a new operation combining features of the Norwood and Rastelli procedures by using what is a natural in situ autograft, the main pulmonary artery as an unobstructed systemic outlet. The left ventricular blood was directed through the VSD into the proximal main pulmonary artery by constructing a tunnel repair as in the Rastelli operation. The VSD, if restrictive, could be enlarged. Then after reconstruction of the aortic arch as for a Norwood operation, the entire systemic circulation was brought over the proximal main pulmonary artery, thus achieving a double-outlet left ventricle with unrestrictive egress of blood from the left ventricle. After this, the right ventricle to pulmonary artery continuity was established with a pulmonary homograft, achieving a biventricular repair. My questions pertain to the very difficult cases of severe hypoplasia of the aortic valve anulus. Were any patients seen in your institution during the same period of time excluded from the type of approach you have taken because of severe hypoplasia of the aortic valve, and what other operations were done in these patients? Also, what role does the preoperative echocardiogram play in assessingthe length of the conal septum, so that
2 9 6 Bove et al.
safe resection of the conal septum can be anticipated. An overestimation of the distance between the aortic valve anulus and the tip of the conal septum could lead to damage of the aortic valve with its catastrophic consequences? Dr. Dominique R. Metras (Marseille, France). Dr. Pridjian and colleagues deserve to be complimented on their success in an extremely difficult subgroup of patients. It seems, however, that the problem of associated subaortic stenosis and aortic valvular hypoplasia in interrupted aortic arch cannot always be adequately treated this way, as several surgeons have, I am sure, sadly experienced. In the literature there are a number of reoperations reported to try to solve this problem. Among these, some have used the VSD and the proximal pulmonary artery for the left ventricular outflow and the associated proximal pulmonary artery to ascending aorta and anastomosis between ascending and descending aorta, and, of course, right ventricular reconstruction with a homograft, as recently reported by Dr. Mee's group in Melbourne. I would like to suggest an alternative procedure that we performed 2 years ago in a neonate with a type B interrupted aortic arch associated with an aberrant right subclavian and severe aortic hypoplasia anulus and a diminutive aortic valve. After transection of the pulmonary artery of the main trunk, excision of subaortic conus, and excision of all ductal tissue, the proximal pulmonary artery was directly anastomosed to the descending aorta, after partial closure of the pulmonary artery due to the different size of both vessels.The VSD was closed, with both vesselsused as left ventricular outflow. Right ventricular outflow was reconstructed with a 13 mm homograft inserted between the right ventricle and pulmonary artery brought anteriorly to the repair with a partial Lecompte maneuver. The patient did well hemodynamically, but, unfortunately, died on day 7 of numerous general and pulmonary infectious problems. We think, however, that this procedure gives a permanent relief of left ventricular obstruction with only one anastomosis at risk, this anastomosis being done without any traction, avoiding complex reconstruction of the aorta. Instead of complex sec-
The Journal of Thoracic and Cardiovascular Surgery February 1993
ondary reoperations such as Konno's procedure or aortic root replacement reported in such cases, the reoperations in this condition would, hopefully, be only conduit replacements. Do you think that the procedure you have suggested is adequate to relieve obstruction in an interrupted aortic arch with diminutive aortic valve and anulus, and if it is not, what kind of operation would you suggest? Dr. John L. Myers (Hershey, Pa.). You deserve congratulations in treating a very difficult group of patients. These patients otherwise are destined to some type of a Norwood procedure and ultimately a Fontan procedure, and I think in these patients with two good ventricles your approach is very appropriate. We have had experience in two similar patients, and in one patient a significant gradient 8 to 9 months postoperatively did subsequently develop. We then treated this much like Konno procedure by placing an incision in the ventricular septum and doing a septoplasty. I think that even if the patient is left with a residual gradient, it can be considered a palliative procedure, and ultimately a Konno operation or septoplasty, if the patient has a good aortic valve and root, will result in a good two-ventricular repair. Dr. Pridjian. In answer to Dr. Chervenkoff and Dr. Metras, I think the advantage of this approach is that it is a direct approach that avoids conduits and hopefully avoids the need for reoperation. In an earlier combined series from Boston and Birmingham, the risk for recurrence of/eft ventricular outflowtract obstruction in patients with interrupted arch and VSD became negligible by 36 months after operation. Our follow-up to this point extends to 20 months in some patients, and we should see fairly soon if reoperation will be necessary. Drs. Chervenkoff, Metras, and Myers all asked questions relating to the issue of the small aortic anulus. The smallest anulus in this group was 4.9 mm and was found in a patient who weighed 1.3 kg. In normal-sized patients, however, an aortoventriculoplasty would ordinarily be required if the aortic anulus was less than 5 to 5.5 mm in diameter provided that the left ventricle was adequate in size.
Edward L. Bove, MD,a L. LuAnn Minich, MDb (by invitation), Ara K. Pridjian, MDa (by invitation), Flavian M. Lupinetti, MDa (by invitation), A. Rebecca Snider, MDb (by invitation), Macdonald Dick II, MDb (by invitation), and Robert H. Beekman III, MDb (by invitation), Ann Arbor, Mich.
h e management of the neonate with malalignment ventricular septal defect (VSD) and aortic arch obstrucFrom the Department of Surgery, Section of Thoracic Surgery; and the Department of Pediatrics, Division of Pediatric Cardiology," The University of Michigan School of Medicine, Ann Arbor, Mich. Read at the Seventy-second Annual Meeting of The American Association for Thoracic Surgery, Los Angeles, Calif., April 26-29, 1992. Address for reprints: Edward L. Bove, MD, 2120 Taubman Center, Box 0344,1500 East Medical Center Dr., Ann Arbor, MI 48109. Copyright © 1993 by Mosby-Year Book, Inc. 0022-5223/93/$1.00/+0.10
12/6/42461
tion is often complicated by the presence of subaortic obstruction resulting from posterior deviation of the infundibular or conal septum. 1 When this obstruction is severe, direct resection by the transaortic approach is difficult because of the small size of the ascending aorta and aortic anulus, which limits the surgical exposure. A number of palliative procedures have been advocated for these patients. They include the use of a left ventricle to descending aorta conduit and anastomosis of the proximal pulmonary artery to the ascending aorta to bypass the obstructed subaortic area.s 3 These procedures are limited by the need for extracardiac conduits and the necessi289
The Journal of Thoracic and Cardiovascular Surgery February 1993
2 9 0 Bove et al.
Table I. Clinical and echocardiographic findings
Patient
Age (days)
Weight (kg)
Aortic arch anatomy
Aortic Anulus size (mm)
Systole
Diastole
28 5 8 22 15 63 14
5.4 3.7 3.6 1.9 1.3 3.4 2.4
CoA CoA lAA,A CoA lAA,B CoA lAA,A
22 ± 20
3.1 ± 1.4
7.0 5.6 6.1 5.3 4.9 6.0 6.8 6.0 ± 0.76
0.65 0.42 0.55 0.49 0.50 0.65 0.46 0.53 ± 0.09
0.73 0.73 0.68 0.56 0.92 0.75 0.71 0.73 ± 0.11
1 2 3 4 5 6 7 Mean
± SD
Ratio LVOT/descending aorta
LVOT, Left ventricular outflow tract; GoA, coarctation; fAA, interrupted aortic arch; A, type A; B, type B; SD, standard deviation.
ty for reoperation. In this condition, however, there remains another surgical approach to the left ventricular outflow tract: through the right atrium and the VSD itself.4, 5 Exposure is generally satisfactory even in the small neonatal heart and may providea more optimal way of directly resecting the obstructing muscle. Because of the limitations imposed by transaortic resection and the use of extracardiac conduits, in September 1989, for patients undergoing VSD closure and primary reconstruction of aortic arch obstruction, we adopted the approach of transatrial resection of the left ventricular outflow tract in the presence of posterior deviation of the infundibular septum. The results in all patients undergoing this procedure at the University of Michigan Medical Center were reviewed and form the basis of this report. Patients and methods From September 1989 to November 1991, seven patients with VSD, aortic arch obstruction, and marked posterior deviation of the infundibular (conal) septum underwent complete repair at C. S. Mott Children's Hospital, the University of Michigan Medical Center, with the use of a technique 'that included transatrial resection of the infundibular septum (Table I). Their ages ranged from 5 to 63 days (median 15 days) and weights from 1.3 to 5.4 kg (mean 3.1 kg). Only one patient was older than 1 month of age. All patients underwent complete two-dimensional and Doppler echocardiography, and three patients also underwent cardiac catheterization as part of their preoperative evaluation (see Table I). In each patient the VSD was of the malalignment type with leftward displacement of the infundibular septum into the left ventricular outflow tract (Fig. I). Because of the presence ofthis nonrestrictive defect, preoperative Doppler measurements of left ventricular outflow velocities were normal. Since calculation of the peak instantaneous gradient from the Bernoulli equation does not reflect the severity of subaortic stenosis in this setting, the degree of obstruction was determined by using anatomic measurements from the two-dimensional echocardiogram. From the standard parasternal long-axis view, the left ventricular outflow dimension was measured from the most posterior
edge of the infundibular septum to the most anterior edge of the opposite left ventricular wall in both systole and diastole (Fig. 2). To provide an index for comparison, this dimension was divided by the descending aortic diameter measured in systole and diastole at the level of the diaphragm from the subcostal sagittal view. For all measurements, at least three cardiac cycles were analyzed and the results averaged. In a separate study from this institution, normal diastolic ratios of the left ventricular outflow tract to the descending aorta were determined to be 1.31 ± 0.14 (mean ± standard deviation) and ranged from 1.00 to 1.54 as measured in a group of 15 infants without heart disease," For the seven patients in this report, the diastolic left ventricular outflow tract to descending aortic ratios ranged from 0.56 to 0.92 (mean 0.73 ± 0.11) and were significantly lower than those in the normal group (p < 0.05). As an additional means of validating the severity of the subaortic stenosis in our patients, the systolic left ventricular outflow tract to descending aortic ratios were measured and compared with previously accepted echocardiographic and angiographic criteria. Severe subaortic obstruction has been considered to be present when that ratio was at or below 0.6. 7 By comparison, our patients ranged from 0.42 to 0.65 with a mean of 0.53 ± 0.09. Coarctation of the aorta was present in four patients and interrupted aortic arch in three patients (type A in two and type B in one). A nonrestrictive infundibular VSD was present in each patient. In two patients, slight extension into the inlet septum was noted, and in one the defect extended to the anulus of the pulmonary valve. Five patients had a patent ductus arteriosus. Two-dimensional and Doppler echocardiographic examinations were repeated in all patients after complete repair. The most recent studies were used for analysis and were performed from 7 days to 13 months after operation (mean 67 days). Because the resection was not circular, remnants of the infundibular septum were frequently seen extending into the subaortic area in the parasternal long-axis view while the left ventricular outflow tract appeared widely patent in other views. For this reason, comparisons of postoperative and preoperative ratios were not performed. Instead, the Doppler peak instantaneous gradient calculated from the Bernoulli equation (valid after VSD closure) was used to assess residual left ventricular outflow obstruction. The level of obstruction was determined by observing the location of the initial increase in velocity with use of color and pulsed Doppler techniques. Surgical technique. All procedures were performed using
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 2
Fig. 1. Preoperative left ventricular angiogram in lateral view demonstrates a large VSD with posterior displacement of the infundibular septum. There was severe coarctation of the aorta with hypoplasia of the transverse aortic arch. deep hypothermia and a single period of circulatory arrest ranging from 22 to 68 minutes (mean 50 minutes). A Y connector was placed in the arterial cannula to allow simultaneous perfusion of the ascending and descending aortas during coolingin those patients with interrupted aortic arch. A single dose of dilute blood cardioplegic solution (30 ml/kg) was administered after cardiopulmonary bypass was discontinued, and the arch vessels were occluded during circulatory arrest. In each patient the aortic arch obstruction was reconstructed by direct anastomosis with the use of continuous absorbable sutures. A right atriotomy was then made, and the VSD was exposed through the tricuspid valve(Fig. 3). A retraction suture was placed on the superior rim of the defect to facilitate exposure of the right coronary cusp of the aortic valve, and the infundibular septum was retracted into view. A wedge of septum was then resected until the anulus of the aortic valve was reached (Figs. 4 and 5). In one patient, resection of the infundibular septum resulted in a small laceration of the base of the right coronary cusp of the aortic valve. This was directly repaired with interrupted sutures through the same right atrial approach. The enlarged VSD was then closed with a suitably shaped patch from the right side of the septum (see Figs. 4 and 5). Cardiopulmonary bypass was reestablished, and systemic rewarming was begun. Bypass was discontinued at a rectal temperature of 35° to 37° C. A right ventriculotomy was not employedin any patient.
Results All patients survived operation and were discharged from the hospital. One late noncardiac death occurred in
Bove et al.
29 1
Fig. 2. Preoperative echocardiogram in long-axis view demonstrates the technique used for left ventricular outflow tract measurements. The left ventricular outflow dimension was measured from the most posterior edge of the infundibular septum to the most anterior edge of the opposite left ventricular wall in both systole and diastole. This dimension was divided by the descending aortic diameter measured in the same cardiac cycle at the level of the diaphragm from the subcostal sagittal view. For all measurements, at least three cardiac cycles were analyzed and the results averaged. LV, Left ventricle; RV, right ventricle; LA, left atrium; Ao, aorta. a premature infant with multiple noncardiac problems 3 months after operation. The six late survivors remain clinically well from 3 to 14 months after repair (mean 8 months). Five of the patients require no cardiac medications. All patients are in normal sinus rhythm, and none has evidence of left bundle branch block. Follow-up two-dimensional and Doppler echocardiographic examination results are shown in Table II. Total peak instantaneous left ventricular outflow tract gradients ranged from 0 to 85 mm Hg, with five of the patients demonstrating no residual gradient. The only patient with a significant gradient had no evidence of subaortic obstruction because the increase in outflow velocity was located at the level of the aortic valve. This patient underwent balloon dilatation 13 months after repair, at which time the catheter gradient was found to be 120 mm Hg. This was reduced to 70 mm Hg after dilatation. No patient had significant subaortic stenosis, although mild turbulence was frequently noted beneath the aortic valve in the area of the resection. The only patient in this series with any subaortic gradient (25 mm Hg) was also the smallest patient in the group (1.3 kg), and in this patient resection of the infundibular septum was the most difficult. Residual aortic arch Doppler gradients were present in two patients, while the remaining five patients had no residual obstruction. One patient underwent successful balloon dilation of the aortic anastomosis 5 months after repair. There were no significant residual ventricular sep-
The Journal of Thoracic and Cardiovascular Surgery February 1993
2 9 2 Rove et at.
Infundibular septum -----,_:;'+_
'---------=--------
Fig. 3. Artist's depictiondemonstratesthe surgeon'sviewof the VSD through the right atrium and tricuspidvalve. A traction suture placed in the infundibular septum facilitates the exposure of the aortic valveanulus (inset).
Fig. 4. A wedgeof infundibularseptum is resecteduntil the anulus of the aortic valveis reached (left). The resulting enlarged VSD is then closed with a patch (right). tal defects. Aortic regurgitation by Doppler echocardiographic examination was absent in four patients and trivial in the remaining three.
Discussion In 1971, Van Praagh and associates'' first described the mechanism of obstruction to left ventricular outflow
resulting from posterior displacement of the infundibular or conal septum in neonates with VSD and aortic arch obstruction. This condition results from malalignment between the infundibular septum with the anterior and posterior extensions of the septal band. The infundibular septum inserts to the left of the posterior limb of the septal band rather than between the posterior and anterior
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 2
limbs. Thus a large malalignment VSD results, and muscular obstruction to left ventricular outflow may be present. The pulmonary valve anulus often overrides the VSD into the left ventricle. This situation is the opposite of that seen in tetralogy of Fallot in which the infundibular septum is displaced anteriorly and to the right of the anterior limb of the septal band, resulting in right ventricular outflow tract obstruction. Van Praagh and others 4• 8 postulated that successful complete repair would be possible if the obstructing muscle were resected. As successful operative treatment of patients with VSD and interrupted aortic arch or coarctation became common, the management of clinically significant subaortic obstruction assumed greater importance. Earlier attempts at reconstruction of the aortic arch combined with pulmonary artery banding were abandoned in many centers because of the finding that the muscular left ventricular outflow tract obstruction rapidly progressed, a processaccelerated by the increased afterload imposed by the band.? Ilbawi and associates'" reported 10 patients with interrupted aortic arch and posterior displacement of the infundibular septum. The four patients in their series who underwent pulmonary artery banding and arch repair all died, whereas all six who underwent placement of a graft between the descending aorta and the pulmonary artery proximal to the band survived. When the displacement of the infundibular septum is severe, arch repair and VSD closure alone may also produce poor results because of unrelieved subaortic obstruction. In an attempt to avoid this situation, alternate surgical approaches have been utilized. Placement of a conduit from the apex of the left ventricle to the descending aorta has been performed in the neonate to bypass the left ventricular outflow tract obstruction.' Alternatively, Yasui and colleagues' reported two infants who underwent successful complete repair by insertion of an intraventricular baffle channeling the VSD to the pulmonary valve, combined with anastomosis of the transected end of the main pulmonary artery to the ascending aorta. The aortic arch was repaired with a graft, and a right ventricle to distal pulmonary artery conduit was inserted to complete the repair. All operative approaches that include insertion of conduits from the right or left ventricle have significant limitations imposed by the lack of potential growth and the need for reoperation. When the aortic valve anulus is adequate in size, direct resection of the posteriorly displaced infundibular septum would appear to be the ideal approach to relieve left ventricular outflow tract obstruction. However, the technical limitations imposed by the frequently small size of the ascending aorta and the aortic valve anulus in this condition make transaortic
Bove et ai. 2 9 3
Fig. 5. Appearance of the VSD and displaced infundibular septumas seenin the lateralview (seeFigs. I and 2 for comparison). Position of the traction suture used to expose the aortic valve during resection (upper left); position of the VSD patch (lower right).
resection of the obstructing muscle difficult. 10 The possibility of approaching the subaortic obstruction through the VSD was proposed by Freedom and associates." This technique was successfully applied by DeLeon and his colleagues- in three infants, two of whom had a transatrial approach to the VSD and subaortic area. The benefits of this technique are substantial and include preservation of the native left ventricular outflow tract as well as avoidance of prosthetic grafts and conduits. Prediction of which patients with VSD and posterior displacement of the infundibular septum will require a more aggressive operative approach to relieve potential obstruction to left ventricular outflow is not always accurate. Some patients with mild degrees of subaortic obstruction will be adequately treated with arch repair and VSD closure alone. Measurement of left ventricular to aortic gradients is not a reliable index of the severity of subaortic stenosis when there is a nonrestrictive VSD or patent ductus arteriosus. Thus there exists a need to define criteria that accurately predict those patients in whom significant early left ventricular outflow tract obstruction from preoperative angiographic or echocardiographic dimensions, or both, will develop. Previously accepted angiographic criteria have included a systolic ratio of the
The Journal of Thoracic and Cardiovascular Surgery February 1993
2 9 4 Bove et al.
Table II. Postoperative two-dimensional and Doppler echocardiographic follow-up Patient
Total LVOT gradient (mm Hg)
Aortic arch gradient (mm Hg)
85
0 0 0
I 2 3 4 5
25
55 25
6 7
0 0
0 0
0 0 0
AR
Residual VSD
0 0
0 0
Trivial Trivial
Tiny
0 Trivial
0
0 0
Comments Gradient at valve level Trivial AR preoperatively, gradient subvalvular Aortic valve injury, balloon dilation residual arch gradient Gradient subvalvular, late death from aspiration
Tiny Tiny
LVOT, Left ventricular outflow tract; AR, aortic regurgitation; VSD, ventricular septal defect.
diameter of the left ventricular outflow tract to that of the descending aorta at the level of the diaphragm of 0.6 or less." Iwahara and associates'! used standard echocardiographic parasternal long-axis views to assess the severity of obstruction in a group of patients with aortic arch obstruction and malalignment VSD and compared them with another group with arch obstruction but without a malalignment VSD. They calculated the systolic ratio of the dimension of the left ventricular outflow tract proximal to the aortic valve to that of the ascending aorta immediately distal to the aortic valve. Those with a malalignment VSD had a mean ratio of 0.59 ± 0.09, which was significantly smaller than that of those patients without a malalignment VSD, 1.03 ± 0.11. Furthermore, the operative mortality among those patients with a malalignment VSD was 56% compared with 0% among those without septal malalignment. These authors concluded that the left ventricular outflow tract to ascending aorta ratio was a useful indicator of the severity of the immediate postoperative left ventricular outflow tract obstruction and that it correlated with the operative outcome. Our method is similar to that of Iwahara, although we chose to use the descending aorta in the ratio. Although it may be argued that the descending aorta may not be reliable in the face of aortic arch obstruction, five of our seven patients had descending aortic flow maintained by a large ductus arteriosus. On the other hand, normalizing the left ventricular outflow dimension to the ascending aorta may be even more questionable since there is often hypoplasia of the ascending aorta in this condition. Additionally, we have found that systolic measurements are not as consistently obtained as those measured in diastole. Thus the diastolic dimensions were also measured for each patient and compared with a group of infants without heart dlsease." The normal infants were found to have significantly larger ratios, 1.31 ± 0.14 versus 0.73 ± 0.11 (p < 0.05). Further confirmation of the accuracy of our method in predicting those patients with significant
subaortic stenosis can be found by analyzing the results in three additional patients who underwent operation for malalignment VSD and aortic arch obstruction in our institution during the same time interval of this study but in whom no infundibular septal resection was done. One patient with mild posterior displacement and a diastolic left ventricular to descending aortic ratio of 1.0 had no left ventricular outflow gradient after operation. The other two patients with ratios of 0.73 and 0.89 had gradients of 36 and 30 mm Hg, respectively. Thus we are confidant that all of the patients in this series had severe posterior displacement of the infundibular septum and would be predicted to have a significantly higher operative mortalityor to demonstrate a greater degree of residual postoperative subaortic obstruction in the absence of any infundibular septal resection. This is further substantiated by the work of Goldmuntz, Baffa, and Weinberg,'? who recommended primary repair for patients with VSD and interrupted aortic arch only when the subaortic region was greater than or equal to 4 mm in the long-axis view. By comparison, the corresponding dimension in our patients was 2.9 ± 0.7 mm (standard deviation) and ranged from 2.1 mm to 3.9 mm. Although the technique of resection of the left ventricular outflow tract obstruction through the VSD in these patients is technically easier than the transaortic approach, there remains the potential for serious complications. Exposure of the aortic valve from the right atrium is difficult, particularly in the tiny neonate, and operative injury to the right coronary cusp of the aortic valve (as occurred in one patient in this series) is a definite hazard. The use of a traction suture to evert the septum into view facilitates the exposure and reduces the possibilityof injury. It must be emphasized that our follow-up is short, and there remains a possibility that subaortic obstruction will progress and require additional intervention in the future. A number of previously reported series have stressed the likelihood of this occurrence and the importance of con-
The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 2
tinued follow-up.P'!" However, the virtual complete absence of significant residual subaortic gradients in our patients, all of whom had severe displacement of the infundibular septum, suggests that early recurrence of obstruction is unlikely. The results in this series demonstrate that resection of left ventricular outflow tract obstruction in the neonate with aortic arch obstruction and a malalignment VSD can be effectively accomplished from a transatrial approach. The ability to relieve the obstruction without the use of external conduits is a significant advantage in this condition. We recommend transatrial resection of the infundibular septum when the diastolic left ventricular outflow tract to descending aorta ratio is less than or equal to 1.0 or the systolic ratio is less than or equal to 0.65. Furthermore, the safety and efficacy of this technique appear to justify its application even in patients with milder forms of septal displacement in whom preoperative uncertainty may exist as to the severity of the left ventricular obstruction that may result after repair. REFERENCES I. Moulaert AJ, Bruins CC, Oppenheimer-Dekker A. Anomalies of the aortic arch and ventricular septal defects. Circulation 1976;53:1011-5. 2. Norwood WI, Lang P, Castaneda AR, Murphy JD. Management of infants with left ventricular outflowobstruction by conduit interposition between the ventricular apex and thoracic aorta. J THORAC CARDIOVASC SURG 1983; 86:771-6. 3. Yasui H, Kado H, Nakano E, et al. Primary repair of interrupted aortic arch and severe aortic stenosis in neonates. J THORAC CARDIOVASC SURG 1987;93:539-45. 4. Freedom RM, Bain HH, Esplagas E, Dische R, Rowe RD. Ventricular septal defect in interruption of aortic arch. Am J Cardiol 1977;39:572-82. 5. DeLeon SY, Ilbawi MN, Roberson DA, et al. Conal enlargement for diffuse subaortic stenosis. J THORAC CARDIOVASC SURG 1991;102:814-20. 6. Minich LL, Snider AR, Bove EL, Lupinetti FM. Echo predictors of the need for infundibular wedge resection in infants with aortic arch obstruction, ventricular septal defect and subaortic stenosis. J Am Coil Cardiol 1992; 19:148A. 7. Hossack KF, Neutze JM, Lowe JB, Barratt-Boyes BG. Congenital valvar aortic stenosis: natural history and assessment for operation. Br Heart J 1980;43:561-73. 8. Van Praagh R, Bernhard WF, Rosenthal A, Parisi LF, Fyler DC. Interrupted aortic arch: surgical treatment. Am J CardioI1971;27:200-11. 9. Matitiau A, Geva T, Colan SD, et al. Bulboventricular foramen size in infants with double-inlet left ventricle or tricuspidatresia with transposed great arteries: influenceon initial palliative operation and rate of growth. J Am Coil CardioI1992;19:142-8.
Hove et af.
295
10. Ilbawi MN, Idriss FS, DeLeon SY, Muster AJ, Benson DW Jr, Paul MH. Surgical management of patients with interrupted aortic arch and severe subaortic stenosis. Ann Thorac Surg 1988;45:174-80. II. Iwahara M, Ino T, Nishimoto K, et al. Clinical features of aortic arch anomaly with malalignment ventricular septal defect. Ann Thorac Surg 1989;48:693-6. 12. Goldmuntz E, Baffa JM, Weinberg PM. Evaluation of the subaortic region in interrupted aortic arch. J Am Coli Cardiol 1992;19:233A. 13. Jones M, Barnhart GR, Morrow AG. Late results after operations for left ventricular outflowtract obstruction. Am J Cardiol 1982;50:569-79. 14. Attie F, OvseyevityJ, Buendia A, et al. Surgical results in subaortic stenosis. Int J CardioI1986;11:329-35. 15. Scott WA, Rocchini AP, Bove EL, et al. Repair of interrupted aortic arch in infancy. J THORAC CARDIOVASC SURG 1988;96:564-8. 16. Sell JE, Jonas RA, Mayer JE, Blackstone EH, Kirklin JW, Castaneda AR. The results of a surgical program for interrupted aortic arch. J THORAC CARDIOVASC SURG 1988; 96:864-77.
Discussion Dr. Cristo I. Tchervenkov (Montreal, Quebec, Canada). I believethat in any malformations associated with hypoplasia of the left heart structures there is a continuum in the degree of severity. Beyond the feasibility of resection of the conal septum, the size of the aortic valve anulus may affect the short-term outcome of this procedure in terms of survival and possibly the long-term outcome in terms of recurrence of left ventricular outflow tract obstruction. In addition to that, what should be done for patients who have aortic valvular stenosis as well? At the Montreal Children's Hospital, we have in 1989 dealt with two patients who had severe left ventricular outflow tract obstruction associated with posterior malalignment VSD and interrupted aortic arch. The marked deviation of the conal septum resulted in a subaortic area of about 3 mm and an aortic valve anulus of 4.5 mm in diameter. In these patients we have performed a new operation combining features of the Norwood and Rastelli procedures by using what is a natural in situ autograft, the main pulmonary artery as an unobstructed systemic outlet. The left ventricular blood was directed through the VSD into the proximal main pulmonary artery by constructing a tunnel repair as in the Rastelli operation. The VSD, if restrictive, could be enlarged. Then after reconstruction of the aortic arch as for a Norwood operation, the entire systemic circulation was brought over the proximal main pulmonary artery, thus achieving a double-outlet left ventricle with unrestrictive egress of blood from the left ventricle. After this, the right ventricle to pulmonary artery continuity was established with a pulmonary homograft, achieving a biventricular repair. My questions pertain to the very difficult cases of severe hypoplasia of the aortic valve anulus. Were any patients seen in your institution during the same period of time excluded from the type of approach you have taken because of severe hypoplasia of the aortic valve, and what other operations were done in these patients? Also, what role does the preoperative echocardiogram play in assessingthe length of the conal septum, so that
2 9 6 Bove et al.
safe resection of the conal septum can be anticipated. An overestimation of the distance between the aortic valve anulus and the tip of the conal septum could lead to damage of the aortic valve with its catastrophic consequences? Dr. Dominique R. Metras (Marseille, France). Dr. Pridjian and colleagues deserve to be complimented on their success in an extremely difficult subgroup of patients. It seems, however, that the problem of associated subaortic stenosis and aortic valvular hypoplasia in interrupted aortic arch cannot always be adequately treated this way, as several surgeons have, I am sure, sadly experienced. In the literature there are a number of reoperations reported to try to solve this problem. Among these, some have used the VSD and the proximal pulmonary artery for the left ventricular outflow and the associated proximal pulmonary artery to ascending aorta and anastomosis between ascending and descending aorta, and, of course, right ventricular reconstruction with a homograft, as recently reported by Dr. Mee's group in Melbourne. I would like to suggest an alternative procedure that we performed 2 years ago in a neonate with a type B interrupted aortic arch associated with an aberrant right subclavian and severe aortic hypoplasia anulus and a diminutive aortic valve. After transection of the pulmonary artery of the main trunk, excision of subaortic conus, and excision of all ductal tissue, the proximal pulmonary artery was directly anastomosed to the descending aorta, after partial closure of the pulmonary artery due to the different size of both vessels.The VSD was closed, with both vesselsused as left ventricular outflow. Right ventricular outflow was reconstructed with a 13 mm homograft inserted between the right ventricle and pulmonary artery brought anteriorly to the repair with a partial Lecompte maneuver. The patient did well hemodynamically, but, unfortunately, died on day 7 of numerous general and pulmonary infectious problems. We think, however, that this procedure gives a permanent relief of left ventricular obstruction with only one anastomosis at risk, this anastomosis being done without any traction, avoiding complex reconstruction of the aorta. Instead of complex sec-
The Journal of Thoracic and Cardiovascular Surgery February 1993
ondary reoperations such as Konno's procedure or aortic root replacement reported in such cases, the reoperations in this condition would, hopefully, be only conduit replacements. Do you think that the procedure you have suggested is adequate to relieve obstruction in an interrupted aortic arch with diminutive aortic valve and anulus, and if it is not, what kind of operation would you suggest? Dr. John L. Myers (Hershey, Pa.). You deserve congratulations in treating a very difficult group of patients. These patients otherwise are destined to some type of a Norwood procedure and ultimately a Fontan procedure, and I think in these patients with two good ventricles your approach is very appropriate. We have had experience in two similar patients, and in one patient a significant gradient 8 to 9 months postoperatively did subsequently develop. We then treated this much like Konno procedure by placing an incision in the ventricular septum and doing a septoplasty. I think that even if the patient is left with a residual gradient, it can be considered a palliative procedure, and ultimately a Konno operation or septoplasty, if the patient has a good aortic valve and root, will result in a good two-ventricular repair. Dr. Pridjian. In answer to Dr. Chervenkoff and Dr. Metras, I think the advantage of this approach is that it is a direct approach that avoids conduits and hopefully avoids the need for reoperation. In an earlier combined series from Boston and Birmingham, the risk for recurrence of/eft ventricular outflowtract obstruction in patients with interrupted arch and VSD became negligible by 36 months after operation. Our follow-up to this point extends to 20 months in some patients, and we should see fairly soon if reoperation will be necessary. Drs. Chervenkoff, Metras, and Myers all asked questions relating to the issue of the small aortic anulus. The smallest anulus in this group was 4.9 mm and was found in a patient who weighed 1.3 kg. In normal-sized patients, however, an aortoventriculoplasty would ordinarily be required if the aortic anulus was less than 5 to 5.5 mm in diameter provided that the left ventricle was adequate in size.