Primary definitive repair of type B interrupted aortic arch, ventricular septal defect, and patent ductus arteriosus

J

THORAC CARDIOVASC SURG

82:501-510, 1981

Primary definitive repair of type B interrupted aortic arch, ventricular septal defect, and patent ductus arteriosus Early and late results Since March. /974. eight patients, aged 7 days to 5 months, with type B interrupted aortic arch (IAA). ventricular septal defect (VSD), and patent ductus arteriosus (PDA) were treated at the Columbia-Presbyterian Medical Center and the University of Maryland Hospital. Six of these patients underwent definitive repair utilizing deep hypothermia and circulatory arrest. Correction involved resection (~l all ductal tissue. primary anastomosis of the aortic arch. closure of the foramen o vale , and patch closure ofthe VSD. Infive patients, all arch vessels were preserved and no prosthetic material was used 10 reconstruct the aortic arch. One patient died 48 hours postoperatively of a coagulopathy . All others survived more than 30 days. One patient, 3'/2 months old at repair, had undergone pulmonary artery banding at another institution at 1I days of age .. he died of recurrent respiratory infections 8 months after correction. Three patients are alive and well 3 10 6 years after repair. Two have undergone repeat cardiac catheterization which demonstrated good growth of the anastomosis and no residual gradient, Primary definitive correction of type B lAA with VSD and PDA provides distinct advantages over palliative or other surgical procedures with excellent long-term results.

Anthony L. Moulton, M.D.* (by invitation), Baltimore, Md., and Frederick O. Bowman, Jr., M.D.,** New York, N. Y.

Type B interrupted aortic arch (IAA) with ventricular septal defect (VSD) and patent ductus arteriosus (PDA) usually causes severe congestive failure in early neonatal life. It is a highly lethal lesion and controversy exists about the optimal management of these patients. Most reported "repairs" have involved the use of prosthetic tubes or sacrifice of some of the arch vessels, often with a simultaneous pulmonary artery banding. This report presents our experience with primary From the Department of Surgery, Division of Chest Surgery, Columbia-Prebsbyterian Medical Center, New York, N. Y., and the Department of Surgery, Division of Cardiothoracic Surgery, University of Maryland Hospital, Baltimore, Md. Read at the Sixty-first Annual Meeting of The American Association for Thoracic Surgery, Washington, D. c., May 11-13, 1981. Address for reprints: Anthony L. Moulton, M.D., Division of Cardiac Surgery, University of Maryland Hospital, 22 S. Greene SI., Baltimore, Md. 21201. *Assistant Professor of Surgery, University of Maryland Hospital. **Clinical Associate Professor of Surgery, Columbia-Presbyterian Medical Center.

definitive repair of the aortic arch with division of the PDA and patch closure of the VSD. Clinical material Patients. Since March, 1974, eight infants with type B IAA plus VSD and PDA have been treated at Columbia-Presbyterian Medical Center and the University of Maryland Hospital. All were symptomatic within the first few days of life, but the patients ranged in age from 3 days to 5 months when referred for treatment. The babies weighed from 2.1 to 3.4 kg at the time of presentation. One patient who presented at 3 1/ 2 months of age had been misdiagnosed as having a simple VSD and had undergone pulmonary artery banding at another institution. All were in refractory congestive heart failure and two were ventilator dependent at the time of transfer. Diagnosis and early management. Chest roentgenograms showed massive cardiomegaly. The initial clinical diagnosis was usually a VSD, coarctation, and PDA, although the patients presenting at 3 and 5

0022-5223/811100501+ 10$01.0010 © 1981 The C. V. Mosby Co.

501

The Journal of

502

Moulton and Bowman

Thoracic and Cardiovascular Surgery

Fig. 1. Angiograms in 3-day-old patient demonstrating type B interrupted aortic arch with ventricular septal defect and patent ductus arteriosus. A. Anteroposterior view of left ventriculogram demonstrates sharp cutoff distal to left carotid artery (a r ro w) and massively dilated pulmonary artery. B. Lateral view of left ventriculogram. The VSD, the ascending aorta. and the innominate and left carotid arteries are well visualized, but there is an abrupt interruption of the aortic arch at that point. months had easily palpable distal pulses. Electrocardiograms demonstrated no axis deviation and were variably indicative of right, left, or biventricular enlargement. The diagnosis was confirmed by cardiac catheterization (Fig. I), except in one patient whose critical condition precluded completion of the study; in this instance coarctation was diagnosed . Two patients had an anomalous right subclavian artery arising from the descending aorta. Prostaglandin E, (POE l )* infusions were initiated in the last two patients and produced dramatic improvement in lower extremity perfusion and urinary output (Fig. 2). The condition of one patient was stabilized with POEl for 4 days until the possibility of necrotizing enterocolitis was excluded . Operation. All eight patients underwent emergency operation (at 7 days to 5 months) . One patient with a muscular VSD underwent primary reconstruction of the arch (anastomosis of the ascending to the descending aorta ) with pulmonary artery banding but died in the operating room. In the critically ill patient who had an incomplete catheterization and a preoperative diagnosis of coarctation , an expanded polytetrafluoroethylene (PTFE)t graft was placed between the carotid artery and descending aorta . The baby remained in congestive heart failure and died at attempted definitive correction *The Upjohn Co.. Kalamazoo. Mich. tGore-Tex. registered trademark of w . L. Gore & Associates, Inc., Elkton, Md .

2 weeks later. A third patient died following primary aortic arch anastomosis with sacrifice of the left subclavian artery and tricuspid regurgitation following attempted VSD closure through the atrium . Five patients underwent one-stage definitive repair of their cardiac lesions. All arch vessels were preserved (including the two anomalous right subclavian arteries) and no prosthetic material was used to reconstruct the aortic arch . Their management is the subject of this report (Table I). After surface cooling to 25° to 28° C (nasopharyngeal), a midsternotomy incision was made . Prior to the administration of heparin (as long as the condition of the patient allowed), the entire ascending aorta, innominate artery, and the left carotid were mobilized . Dissection was extended along the ductu s arteriosus to the descending aorta and left subclavian artery as far as possible. Cardiopulmonary bypass was initiated with a single 2 mm metallic cannula in the ascending aorta and a single 20 Fr. venous cannula in the right atrium, and the patient was core-cooled to 14° to 16° C. Satisfactory lowering of rectal temperature was noted and urine output remained adequate . (In one patient a femoral arterial line documented good perfusion of the lower part of the body . In another with an anomalous right subclavian artery arising from the descending aorta , radial artery pressures were indicative of satisfactory distal perfusion .) During this time, mobilization of the ascending aorta, ductus, and descending aorta was completed (Fig. 3). Bypass was then terminated and the

Volume 82 Number 4

Repair of type B fAA with VSD and PDA

October. 1981

5 03

LSCA

LV

pressure

PRE - PGE1

POST - PGE1

Fig. 2. Pressure tracings in a 9-day-old patient with type B interrupted aortic arch with ventricular septal defect and patent ductus arteriosus. Dramatic improvement in the pressure and pulse contour in the descending aorta is seen minutes after the initiation of prostaglandin E! (POE!) infusion, with almost complete abolition of the pressure gradient between the left ventricle (LV) and descending aorta.

patient was exsanguinated via the venous line. The venae cavae and the innominate and carotid arteries were clamped with vascular bulldog clamps. The cannulas were removed. In the last patient, cardioplegic solution was infused into the aortic root. The ductus was transected at its junction with the descending aorta, and all ductal tissue was excised back to clearly healthy aortic tissue. The vagus nerve and its recurrent laryngeal branch were carefully identified and preserved. In two patients, the incision was carried into the proximal subclavian artery to guarantee a wide orifice. An incision was then made in the posterolateral ascending aorta and extended into the proximal posterior left common carotid artery (Fig. 4). An end-to-side anastomosis between the descending and ascending aorta was then created with continuous 6-0 and 7-0 Prolene* suture. This was easily accomplished without obvious tension. In none of these patients was it necessary to sacrifice any intercostal or arch vessels to achieve satisfactory approximation of the vessel. A short incision in the lateral right atrium was made and the foramen ovale sutured. The ventricular anatomy was inspected through the tricuspid valve, but in each of these patients a short right ventriculotomy was necessary to visualize the VSD adequately. The incision *Ethicon, Inc., Somerville, N. J.

Fig. 3. Type B interrupted aortic arch with ventricular septal defect and patent ductus arteriosus. Note discrepancy between aortic and pulmonary artery diameters and the great distance between ascending and descending aorta. RSCA. Right subclavian artery. LSCA. Left subclavian artery. RCA. Right carotid artery. LCA. Left carotid artery. PDA. Patent ductus arteriosus. MPA. Main pulmonary artery. AO. Aorta. VSD. Ventricular septal defect.

RSCA

.~ '\VSD "

Fig. 4. All identifiable ductal tissue has been excised. The incision in the distal ascending aorta has been extended into the carotid artery to allow better approximation and the widest possible anastomosis. If necessary, the incision in the descending aorta can be extended into the subclavian artery. For abbreviations see Fig. 3.

The Journal of

5 04

Moulton and Bowman

Thoracic and Cardiovascular Surgery

RSCA

RCA

Table I. Summary of overall surgical experience with type B interrupted aortic arch with ventricular septal defect and patent ductus arteriosus since March,1974

~ 1974 1974 1974 1975 1977 1977

Age

3 rno I rno 5 rno 3 rno II days 2 rno

1978 7 days 1980 9 days

Operation

Definitive Partial Definitive Definitive Definitive Definitive Definitive I. Partial 2. Definitive

Result

Late death-8 rno Operative death-PA banding Living and well Living and well Operative death-LCO Operative deathcoagulopathy Living and well Persistent CHF Operative death-LCO

Legend: PA. Pulmonary artery. LCO. Low cardiac output. CHF. Congestive heart failure.

Fig. 5. The completed repair. The ventricular septal defect has been patched via a ventriculotomy and the foramen ovale closed. An end-to-side anastomosis with a single running suture has been created. and the pulmonary artery (PA) has been repaired. For abbreviations see Fig. 3.

was transverse in two and vertical in three patients. The VSD was patched with Teflon felt by means of reinforced mattress sutures. Cardiopulmonary bypass was reinstituted and air was carefully evacuated from the left ventricle and ascending aorta before the carotid clamps were released. The average circulatory arrest time was 76 minutes (range 62 to 88). The left ventricle was not vented during rewarming. All remaining ductal tissue was resected from the pulmonary artery. In two patients, the pulmonary artery was so massively dilated that a wedge of redundant tissue was excised from its anterior wall. In another, a "reefing" suture was utilized to decrease the pulmonary artery diameter (Fig. 5). When rewarming was complete, the patient was weaned from bypass. Right ventricular pressures measured following the repair were 30 to 35 mm Hg, with right atrial pressures of 9 to 12 mm Hg. Simultaneous systemic pressures were 85 to 90 mm Hg, with left atrial pressures of 5 to 8 mm Hg. Results Early deaths. All five patients survived the corrective surgical procedure. Although hemodynamically stable, one 2Vi-month-old patient died of a coagulopathy 48 hours postoperatively. Autopsy confirmed diffuse intraparenchymal hemorrhage, although extensive endocardial fibroelastosis of all

chambers was also noted. The surgical repair was satisfactory. The 31/z-month-old baby who previously had undergone pulmonary artery banding had a protracted hospital course. He required reoperation for tamponade and had hypocalcemic seizures, multiple episodes of sepsis, and gastrointestinal bleeding. Six weeks postoperatively he underwent intestinal resection for obstruction from adhesions from healed enterocolitis. Thereafter he was hemodynamically stable, but he had multiple episodes of pneumonia and eventually died 8 months following correction. Postmortem examination again showed a dilated right ventricle with endocardial fibroelastosis, severe chronic pneumonia, and diffuse gastritis. The surgical repair was intact with the aortic anastomosis widely patent. Long-term follow-up. There are three long-term survivors. Two patients are alive and well 3 to 6 years postoperatively. The third patient has been lost to follow-up but was well when seen 2 years ago (5 years postoperatively), despite early postsurgical concerns about left bronchial compression. Two of these patients have undergone postoperative catheterization. The baby operated upon at 7 days of age was studied 8 months later and appears to have a stenosis in the new transverse arch (Fig. 6). Pressures in the ascending and descending aorta were equal, however, and the narrowing lies at the proximal base of the carotid artery, while the actual anastomotic site between the carotid and subclavian arteries appears widely patent. There was a 15 mm Hg gradient across the bicuspid aortic valve. Mild subaortic stenosis is also suggested in the angiogram on this patient, but again no gradient could be demonstrated in this area. Ventricular size and function were normal, and there were no residual shunts. A 3 year follow-up physical examination

Volume 82 JUnber 4 CklDber. 1981

Repair of type B fAA with VSD and PDA

505

Fig. 6. Angiograms obtained 8 months following repair of type B interrupted aortic arch with ventricular septal defect and patent ductus arteriosus at 7 days of age. A . An anteroposterior left ventriculogram shows no residual ventricular shunt and good apposition of the carotid and subclavian arteries. The aortic contour appears normal in this view. B. A lateral left ventriculogram in the same patient demonstrates mild stenosis at the proximal base of the left carotid, most likely from angulation of that vessel. The anastomotic site between the carotid and subclavian arteries appears widely patent.

Fig. 7. Angiograms of another patient 2'h years following definitive repair. A . This anteroposterior left ventriculogram demonstrates complete closure of the ventricular septal defect and a normal-appearing aortic arch. The right subclavian artery arises anomalously from the descending aorta. B. The lateral left ventriculogram demonstrates a normal-appearing aortic arch, confirming growth at the anastomotic site. shows a 30 mm Hg differential blood pressure between the upper and lower extremities. The second patient was 3 months old at correction and was recatheterized 2'h years later (Fig . 7) . This study demonstrated an intact ventricular septum with no shunts, normal arch vessels, and no narrowing or gradient between the ascending and descending aorta. These observations document growth of the anastomosis. Six years after the operation, she is nor-

motensive, athletically active , and continuing to do well.

Discussion IAA is a rare lesion accounting for approximately I % of cases of congenital heart disease . However, it is highly lethal and accounts for 4% of deaths from heart disease within the first month of life. 1-3 In a recent review, Trusler and Freedom" noted that 72% of their

The Journal of

506

Moulton and Bowman

Thoracic and Cardiovascular Surgery

A

42%

C

B

53 %

4%

Fig. 8. Classification of aortic arch interruptions according to Celoria and Patton." Type A. The interruption lies distal to the left subclavian artery . Type B. The interruption lies between the left carotid and subclavian arteries . leaving a greater distance between the aortic divisions than in A or C. Type C. The interruption is between the innominate and carotid arteries. The percentages listed below are the incidence of each type as found in a review of the literature by Van Praagh and associates" in 1971 and corroborated by several other studies since .

Fig. 9. Operative photograph in a 9-day-old , 2.4 kg baby demonstrating exposure via a left thoracotomy . The snare is around the left carotid artery . Note the discrepancy between the diameters of the descending aorta and the carotid and subclavian arteries . The vagus nerve and its recurrent laryngeal branch are clearly seen . This patient was misdiagnosed as having a simple coarctation of the aorta and underwent placement of a PTFE graft between the carotid artery and descending aorta . The baby died at subsequent intracardiac repair.

patients with IAA presented within the first week. Untreated, the median age at death is 4 to 10 days. The first month's mortality for the lesion without therapy is 76 %, and 90% will die within the first year. 2. s , 6 IAA is rarely an isolated lesion, and its presentation

and prognosis are closely related to the associated lesion. A PDA is an almost universal finding, and flow to the distal aorta is dependent on it. A VSD is present in over 90% of cases, and truncus arteriosus or aortopulmonary window are frequently seen. I. 2 . 4 A bicuspid aortic valve is present in one third to one half of patients." 6. 7 Significant subvalvular obstruction may be present as well, caused by ventriculoseptal malalignment, abnormal mitral attachment, or an anterolateral muscle bundle of left ventricle. 8 A truly hypoplastic left heart is rarely associated, although the normal-sized left ventricle may appear small in contrast to a dilated, hypertrophied right ventricle. Absent thymic tissue is frequent (l0% in the series of Collins-Nakai and assoelates") , and all these neonates should be screened for DiGeorge 's syndrome (III-I V pharyngeal pouch deficiency) ." 9. 10 If present , hypocalcemia and immunologic problems must be treated, and only irradiated blood must be transfused. In type B IAA , the right subclavian artery frequently arises from the descending aorta, as seen in two of the eight patients in this series. In all three types of IAA classified by Celoria and Patton 11 (Fig . 8), flow to the descending aorta is ductus dependent. As the PDA starts to close, usually within the first few days of life, rapid deterioration follows . Perfusion of the lower part of the body and urinary output become inadequate. As the ductus constricts, blood that is being shunted left to right through an associated intracardiac lesion (particularly if there is concomitant subvalvular or valvular aortic stenosis) is then forced into the pulmonary circulation . This in-

Volume 82

Repair of type B fAA with VSD and PDA

Number 4 October, 1981

°

5 7

Table II. Published results of one-stage definitive repair of type B interrupted aortic arch with ventricular septal defect and patent ductus arteriosus Ref. No.

No. of patients

23

I

25

3 4 6 5

29 27* Present series Totals

19

Early death

Late death

2

2 I I 6

I I 2

(31%)

(11%)

Long-term survivors

Duration of survival (yr)

I I

9 8

2

2-4 1-5

4 3 II (58%)

3-7 1-9

*J L Monro. personal communication.

creases left-sided volume overload, which eventually leads to rapidly progressive biventricular failure. Diuretics do not help because of inadequate renal perfusion, and metabolic acidosis develops. Without treatment, the patient dies. The operation for various types of IAA has a long history. The first successful surgical repair of type B IAA with VSD and PDA is generally attributed to Merrill, Webster, and Samson " (1957), who divided the PDA and anastomosed its proximal end to the ascending aorta in a 3V2-year-old child. The two VSDs were successfully closed 4 years later. That same year, Mustard!" had divided the PDA and anastomosed the ascending to the descending aorta in a 7 1/2-month-old infant with type A IAA. Successful closure of the VSD was performed 3 1/2 years later, but the patient died of atrioventricular block 8 months afterward. In 1959, Quie and associates':' accomplished aortic arch reconstruction in a 4 1/2-month-old infant with type B IAA with VSD and PDA by dividing the PDA and the left carotid artery and performing an end-to-end carotid artery-descending aorta anastomosis. The baby died 2 1/2 days later with a mucus plug. Blake, Manion, and Spencer" (1962) reported the first use of a prosthetic graft between the ascending and descending aorta in a 20-month-old boy with type B IAA. The ductus was divided, but the VSD was left open. In the neonatal group, Tyson and associates 16 (1970) performed a successful palliative operation in a 12day-old infant with type A IAA by end-to-end subclavian artery-descending aorta anastomosis combined with pulmonary artery banding. That same year, the Boston Children's Hospital group'? palliated an 11day-old infant with type B IAA with a conduit from the main pulmonary artery to the descending aorta plus pulmonary artery banding; this approach later was supported by Jones, Plauth, and Hatcher!" (1975) but subsequently abandoned following disappointing longterm results. An II-day-old patient operated upon by

Ventemiglia and colleagues'" in 1970, who performed direct aortic arch reconstruction and pulmonary artery banding, underwent VSD closure and debanding in 1973 and was thriving at least 3 years later. This patient may well be the first long-term survivor of an operation for type B IAA with VSD and PDA to undergo the operation in the neonatal period. Successful completion of both stages of repair of type B IAA with VSD and PDA in infancy was performed by the group at the University of California in San Francisco in 1973, though not reported until 1976. 20 At 2 days of age, a left carotid-descending aorta anastomosis was created, followed by VSD closure at 6 months. The baby was well 3 years later. Repair of type B IAA associated with type I truncus arteriosus had been achieved by Gomes and Mcfloon'" in a 2-year-old child in 1971. Simultaneous definitive repair of type B IAA with VSD was reported in 1976 by Singh, Bentall, and Oakley" in a 6-year-old child. The first successful one-stage correction of type B IAA with VSD and PDA in infancy was described by Murphy and associates'" in 1973, although the preceding year Barratt-Boyes and co-workers" had reported simultaneous repair (with a 12 mm Dacron conduit) in an 8-day-old infant with type A IAA with VSD, PDA and intracardiac total anomalous pulmonary venous drainage under deep hypothermia and circulatory arrest. Murphy's repair in a 3-day-old infant involved use of a portion of the basilic vein from the baby's father to reconstruct the aortic arch, and the child is still doing well. * The atrial septal defect and VSD were closed with pledget-supported mattress sutures. Barratt-Boyes had utilized a left posterolateral thoracotomy and midsternotomy for his repair; Murphy extended his midsternotomy incision into the left third intercostal space for better exposure ofthe descending aorta. In 1976, Trusler and Izukawa" demonstrated that primary aortic arch *ARC Dobell, personal communication, 1981.

508

The Journal of

Moulton and Bowman

Thoracic and Cardiovascular Surgery

reconstruction for correction of type B IAA with VSD and PDA could be achieved by a standard midsternotomy alone, a finding corroborated by our own group that same year. 26 Through a midsternotomy in 1977, Monro, Brown, and Conway-" repaired this anomaly in a 6-month-old child by creating a tube from the proximal pulmonary artery with the PDA intact and anastomosing it to the ascending aorta. He has since successfully operated upon three others (3 days to 7 months old), although in the last two neonates he has utilized the proximal left carotid to reconstruct the aortic arch and has implanted the distal left carotid into the right carotid. Postoperative studies demonstrate good growth of the new carotid "arch."* Bailey and co-workers'" operated upon a 2-day-old neonate in 1978 via a bilateral anterior thoracotomy, an approach they-" have used successfully in two subsequent cases with excellent 18 month follow-up. Bailey, however, mobilizes 75% to 80% of the descending aorta by cauterizing the intercostals, a technique that we have not found necessary and that carries potential risks, although none has been noted yet. The overall experience with one-stage repair by several groups is summarized in Table II. The introduction of POE, to dilate the PDA and prevent its closure has dramatically changed the outlook for these neonates.:" 32 Infusion of POE, can significantly improve perfusion in the descending aorta (see Fig. 1) and allow diuresis and reversal of acidosis. Although emergency operation is still necessary, POE, allows stabilization of the patient during diagnostic procedures and allows the operation to be performed under optimal conditions. One patient in this series was maintained on POE, for 4 days until the possibility of necrotizing enterocolitis was excluded. These critically ill neonates require immediate diagnostic studies, which must be complete to allow planning for the appropriate surgical procedure. The correct diagnosis was not made preoperatively in one patient in this series or in eight of the 41 patients in the Toronto series reported by Esplugas. 33 If the patient's condition permits, an ascending aortogram should be performed, since the massive shunt in a left ventriculogram highlights a dilated pulmonary artery and obscures the aortic anatomy. Aortography may also demonstrate subclavian steal, as seen in seven of the patients reported by Esplugas. 33 The intracardiac anatomy should be clearly delineated and a truncus arteriosus and aortopulmonary window should be excluded. Any subaortic obstruction should also be demonstrated. The anatomy of the associated VSD is of more than academic interest. In a careful review of 34 specimens

of IAA by Freedom and associates," only five had a classical membranous VSD, three had an atrioventricular canal type of defect, and three had a muscular VSD. The majority, 21 (62%), had a defect in the conal septum, with four being directly subpulmonary. Most are associated with some conoventricular malalignment and some degree of subaortic stenosis. The therapeutic ramifications are twofold. As postulated by Van Praagh and colleagues" in 1971 and confirmed by Trusler and Freedom" in 1979, any procedure which may result in further hypertrophy of the leftward deviated conal septun, for example pulmonary artery banding, may create or aggravate subaortic stenosis. Therefore, in the presence of a conoventricular malalignment type of VSD, pulmonary artery banding would appear to be contraindicated and primary definitive repair would be the treatment of choice. At the time of repair, the ventricular anatomy should be inspected through the tricuspid valve, but a ventriculotomy will frequently be required to expose the margins of these high VSDs. The distance between the two ends of the interrupted aorta are greatest in type B (see Fig. 8). Early successes with palliative procedures utilizing prosthetic tubes, particularly since the advent of POE" have been reported." but long-term follow-up reveals few survivors. Though the technique has been useful in type A IAA, it does not appear to work as well in bridging the greater gap in type B. A small conduit also exposes the patient to the risks of reoperation. Similarly, the subclavian and carotid arteries are often not of adequate caliber to alleviate the obstruction completely" (Fig. 9), though some successes with their use (most notably Monro's recent cases) have been achieved. Exposure of the ascending aorta to allow placement of a larger conduit is difficult through a left thoracotomy. Even if exposed, the aorta is usually of small diameter, and partial occlusion is either poorly tolerated or impossible. Copeland and associates" reported the use of partial cardiopulmonary bypass for a palliative procedure in a critically ill, acidotic neonate. An infusion of POE, might have allowed stabilization, but if cardiopulmonary bypass is required, it would seem more advantageous to proceed with definitive repair. Monro* reports satisfactory follow-up in two patients, 5 and 7 months old at repair, in whom ductal tissue was used to reconstruct the aortic arch. However, ductal tissue has the potential for constriction, and we would be reluctant to employ any procedure which routinely utilized the PDA as part of the arch repair. 37 It is interesting to speculate whether the presence of an anomalous right subclavian artery allows greater

*lL Monro, personal communication, 1981.

*lL

Monro, personal communication, 1981.

Volume 82 Number 4 October, 1981

mobilization of the ascending aorta, so that there is less tension on a primary anastomosis in the arch construction. An anomalous right subclavian artery was present in the patients treated successfully by Murphy, 23 Trusler;" Bailey, 29 and their associates and in the survivor in Monros-" last three patients. An anomalous right subclavian artery was present in two of the five patients undergoing definitive correction in the present series, including the patient who is thriving 6 years postoperatively with the normal-appearing postoperative angiogram. Obviously, the overall mortality for any approach to IAA precludes any statistically significant conclusions. However, the 58% long-term survival ranging from 1 to 9 years after one-stage correction (Table II) is superior to the long-term results obtained with other approaches. Because of our experience, we believe that primary definitive correction of type B IAA with VSD and PDA can be achieved reproducibly, with preservation of all arch vessels but without the need for prosthetic materials, by following these steps: (I) median sternotomy and partial bypass (cannulas in ascending aorta and right atrium only) performed with the aid of profound hypothermia and total circulatory arrest; (2) complete resection of all ductal tissue; (3) primary anastomosis between ascending and descending aorta (including left carotid and subclavian arteries if necessary for wide anastomosis); and (4) VSD closure via a ventriculotomy. This procedure offers definite advantages over palliative or other surgical procedures and provides excellent long-term results. Addendum Since this report was submitted, another neonate with type B IAA with type III truncus arteriosus has been treated at the University of Maryland Hospital. Anuric and moribund on arrival, she improved on prostaglandin infusion. At 5 days of age primary anastomosis of the aortic arch and repair of the pulmonary artery were performed. Although this procedure was easily accomplished with only 18 minutes of circulatory arrest, it was elected to band both pulmonary arteries (preoperatively a type I truncus had been anticipated and definitive repair had been planned). The baby was weaned from bypass uneventfully with equal radial and umbilical artery pressures and acceptable Pojs, but she remained in renal failure and died of sepsis on the third postoperative day. REFERENCES Moller JH, Edwards JE: Interruption of the aortic arch. Anatomic patterns and associated cardiac malformations. AJR 95:557-572, 1965 2 Van Praagh R, Bernhard WF, Rosenthal A, Parisi LF, Fyler DC: Interrupted aortic arch. Surgical treatment. Am J CardioI27:200-211, 1971

Repair of type B fAA with VSD and PDA

509

3 Collins-Nakai RL, Dick M, Parisi-Buckley L, Fyler D, Castaneda AR: Interrupted aortic arch in infancy. J Pediatr 88:959-962, 1976 4 Trusler GA, Freedom RM: Surgical approach to the management of interruption of the aorta, vol 2, chap 20, MJ Godman, RM Marquis, eds., Edinburgh, 1979, Churchill Livingstone, p 268-273 5 Roberts WC, Morrow AG, Braunwald E: Complete interruption of the aortic arch. Circulation 26:39-59, 1962 6 Freedom RM, Bain HH, Esplugas E, Dische R, Rowe RD: Ventricular septal defect in interruption of aortic arch. Am J Cardiol 39:572-582, 1977 7 Immagoulou A, Anderson RC, Moller JH: Interruption of the aortic arch. Clinical features in 20 patients. Chest 61:276-282, 1972 8 Moulaert AJ, Bruins CC, Oppenheimer-Dekker A: Anomalies of the aortic arch and ventricular septal defects. Circulation 53:1011-1015, 1976 9 Harvey JC, Dungan WT, Elders MJ, Hughes ER: Third and fourth pharyngeal pouch syndrome, associated with vascular anomalies and hypocalcemic seizures. Clin Pediatr 9:496-499, 1970 10 Freedom RM, Rosen FS, Nadas AS: Congenital cardiovascular disease and anomalies of the third and fourth pharyngeal pouch. Circulation 46: 165-172, 1972 II Celoria GC, Patton RB: Congenital absence of the aortic arch. Am J Heart J 58:407-413, 1959 12 Merrill DL, Webster CA, Samson PC: Congenital absence of the aortic isthmus. J THORAC SURG 33:311-316, 1957 13 Mustard WT: Cited by Van Praagh et al 2 14 Quie PG, Novick R, Adams P, Anderson RC, Varco RL: Congenital interruption of the aortic arch. J Pediatr 54:87-92, 1959 15 Blake HA, Manion WC, Spencer FC: Atresia or absence of the aortic isthmus. J THoRAc CARDIOVASC SURG 43:607-611,1962 16 Tyson KRT, Harris LC, Ngiem QX: Repair of aortic arch interruption in the neonate. Surgery 67: 1006-1010, 1970 17 Litwin SB, Van Praagh R, Bernhard WF: A palliative operation for certain infants with aortic arch interruption. Ann Thorac Surg 14:369-375, 1972 18 Jones EL, Plauth WH, Hatcher CR Jr: A palliative operation for all types of aortic arch interruption in the neonate. J THoRAc CARDIOVASC SURG 69:579-589, 1975 19 Ventemiglia R, Oglietti J, Wukasch DC, Hallman GL, Cooley DA: Interruption of the aortic arch. J THORAC CARDIOVASC SURG 72:235-242, 1976 20 Fishman NH, Bronstein MH, Berman W Jr, Roe BB, Edmunds LH Jr, Robinson SJ, Rudolph AM: Surgical management of severe aortic coarctation and interrupted aortic arch in neonates. J THORAC CARDIOVASC SURG 71:35-48, 1976 21 Gomes MM, McGoon DC: Truncus arteriosus with interruption of the aortic arch. Report of a case successfully repaired. Mayo Clin Proc 46:40-43, 1971 22 Singh MP, Bentall HH, Oakley CM: Successful total correction of congenital interruption of the aortic arch and

The Journal of

5 10

23

24

25

26 27

28

29

30

31

32

33

34

35 36

37

Moulton and Bowman

ventricular septal defect. Thorax 25:615-623, 1970 Murphy DA, Lemire GG, Tessler I, Dunn GL: Correction of type B aortic arch interruption with ventricular and atrial septal defects in a three-day-old infant. J THORAC CARDIOVASC SURG 65:882-886, 1973 Barratt-Boyes BG, Nichols TT, Brandt PWT, Neutze JM: Aortic arch interruption associated with patent ductus arteriosus, ventricular septal defect, and total anomalous pulmonary venous connection. J THORAC CARDIOVASC SURG 62:367-373, 1972 Trusler GA, Izukawa T: Interrupted aortic arch and ventricular septal defect. Direct repair through a median sternotomy incision in a 13-day-old infant. J THORAC CARDIOVASC SURG 69:126-131,1975 Bowman FO Jr: Discussion of Fishman et aFo Monro JL, Brown W, Conway N: Correction of Type B interrupted aortic arch with ventricular septal defect in infancy. J THORAC CARDIOVASC SURG 74:618-623, 1977 Monro JL: Interrupted aortic arch-Techniques and results, Congenital Heart Disease Under Three Months of Age, Bologna, Italy, 1981, Patron Editore, pp 375-382 Bailey LL, Jacobson JG, Vyhmeister E, Petry E: Interrupted aortic arch complex. Successful total correction in the neonate. Ann Thorac Surg 25:66-70, 1978 Bailey L, Jacobson J, Doroshow R, Merritt W, Petry E: Anatomic correction of interrupted aortic arch complex in neonates. Proc First World Congress on Paediatric Cardiology, London, 1980 Heyman MA, Berman WB, Rudolph AM, Whitman V: Dilation of the ductus arteriosus by prostaglandin E 1 in aortic arch abnormalities. Circulation 59: 169-173, 1976 Radford DT, Bloom KR, Coceani F, Fariello R, Olleg PM: Prostaglandin E 1 for interrupted aortic arch in the neonate. Lancet 2:95, 1976 Esplugas E: Interruption of the aorta-Anatomical, clinical and angiocardiographic observations, Paediatric Cardiology, vol 2, Heart Disease in the Newborn, MJ Godman, ed., New York, 1979, Churchill Livingstone, pp 187-195 ZOOka KG, Roland JMA, Cutiletta AF, Gardner TJ, Donahoo JS, Kidd L: Management of aortic arch interruption with prostaglandin E 1 infusion and microporous expanded polytetrafluoroethylene grafts. Am J Cardiol 46:1001-1005, 1980 Allard JR, Williams PL, Dobell ARC: Interrupted aortic arch. Ann Thorac Surg 21:243-256, 1975 Copeland JG, Record JA, Salomon NW, Soon DJ, Allen HD, Goldberg SJ: Successful palliation using partial cardiopulmonary bypass in a two-day-old infant with type B interruption of the aortic arch. J THORAC CARDIOVASC SURG 76:495-499, 1978 Muraoka R, Yokota M, Aoshima M, Nomoto S, Osaraqi M, Kyoku I, Nakano H, Ueda K, Saito A: Simplified method for total correction of interrupted aortic arch with ventricular septal defect in infancy. J THORAC CARDIOVASC SURG 78:744-749, 1979

Thoracic and Cardiovascular Surgery

Discussion DR. GEORGE A. TRUSLER Toronto, Ontario, Canada

The operation described by Dr. Moulton is not easy to perform, but it can be done. The key is to operate early, while the infant is in good condition. Most of the dissection should be done before going on bypass, so that once on bypass the surgeon can proceed with the repair. These children are so ill that if they have a cardiac arrest before bypass, they probably will not survive. After our first case, we operated upon several children unsuccessfully and upon one who did well. For a child in good condition, this is an effective operation. DR. BOW MAN (Closing) Our major problem with these infants has been with postoperative bleeding. For this reason we have used continuous suture at the line of anastomosis. Once this anastomosis is accomplished and the vascular tree is refilled, you no longer can see it in case there is any suture line bleeding. Actually, one of our patients did die of a coagulopathy and in others we did have coagulation problems which we overcame with time and effort. We believe that in our hands definitive correction is the best procedure. Dr. Trusler taught us how to do this operation, and we strongly recommend it as the procedure of choice for these patients. DR. M 0 U L TON (Closing) I thank Dr. Trusler and Dr. Bowman for their comments. When we have discussed the one-stage procedure with others who have advocated palliative procedures, the argument has always been that this group of patients is so critically ill that they will not tolerate bypass. In contrast, I would like to emphasize the limited exposure through the left thoracotomy. In the patient who was misdiagnosed as having a coarctation, there was adequate visualization of the massively dilated ductus arteriosus, the descending aorta, and the subclavian artery (see Fig. 9). The carotid artery can be brought into the field, but the huge pulmonary artery precludes adequate exposure of the ascending aorta, which is usually hypoplastic. In this patient, we elected to do a palliative procedure, since the thoracotomy incision had already been made. A 4 mm PTFE graft was placed between the carotid artery and the descending aorta. It would be hard to imagine getting a much larger conduit in a baby of this size. Constriction of the ductus arteriosus resulted in a precipitous drop in umbilical artery pressure despite a wide open anastomosis on re-examination. We had hoped that the size of the proximal subclavian artery would increase to provide increased flow through the graft and allow later ductal ligation. Still in failure 2 weeks later, this baby died at subsequent correction with a marked differential blood pressure between the upper and lower parts of the body. I think it is hard to base palliation in these babies on these tiny arch vessels. They cannot survive inadequate perfusion of the lower part of the body.