Biventricular Repair With the Yasui Operation (Norwood/Rastelli) for Systemic Outflow Tract Obstruction With Two Adequate Ventricles

Transcription

1 Biventricular Repair With the Yasui Operation (Norwood/Rastelli) for Systemic Outflow Tract Obstruction With Two Adequate Ventricles Kirk R. Kanter, MD, Paul M. Kirshbom, MD, and Brian E. Kogon, MD Division of Cardiothoracic Surgery, Department of Surgery, Emory University School of Medicine, and Pediatric Cardiac Surgery, Children s Healthcare of Atlanta at Egleston, Atlanta, Georgia Background. The Yasui procedure is employed in neonates with interrupted aortic arch and left ventricular outflow tract obstruction (IAA/LVOTO) or aortic atresiasevere stenosis with ventricular septal defect (AA/VSD) and 2 adequate-sized ventricles. This combines a Norwood arch reconstruction with a Rastelli operation establishing a biventricular repair. Methods. From 2002 to 2011, 21 neonates aged 3 to 55 days (mean 12.2 days, median 7 days) had IAA/LVOTO (n 13), AA/VSD (n 7), or AA/IAA with aortopulmonary window (n 1); ten (48%) had genetic abnormalities (8 with DiGeorge syndrome). Based on clinical characteristics and surgeon preference, 6 had a primary Yasui repair (4 AA/VSD, 2 IAA/LVOTO); 15 were staged with an initial Norwood repair (3 AA/VSD, 12 IAA) followed by Yasui completion in 13 (2 await completion) 4.3 to 26.6 months later (median 6.9 months). Results. Early mortality was zero with no interstage deaths in the staged patients. One patient died 2 months after staged repair. Since biventricular repair, 8 survivors (44%) had reoperation for conduit replacement (n 6), recurrent LVOTO (n 1), or a residual VSD (n 1). No patient requires a pacemaker. There were 3 late deaths after biventricular repair, all in patients with genetic syndromes and IAA/LVOTO. Actuarial survival after initial operation was 100% at 1 year and 75% at 5 years. Actuarial freedom from reoperation or death after biventricular repair was 14% at 5 years. Conclusions. The Yasui operation is effective for patients with IAA/LVOTO and AA/VSD. Primary and staged repair have comparable results. Reoperation after biventricular repair seems inevitable, mostly for conduit replacement. Genetic factors may affect longterm survival. (Ann Thorac Surg 2012;93: ) 2012 by The Society of Thoracic Surgeons Interrupted aortic arch and severe left ventricular outflow tract obstruction (IAA/LVOTO) or aortic atresia with a ventricular septal defect (AA/VSD) and 2 wellformed ventricles presents surgical challenges. An early success was a staged approach described by Norwood and Stellin in 1981 [1] for a newborn with interrupted aortic arch, VSD, and AA. Initially, a valved conduit from the left ventricle to the descending aorta was placed with an 8-mm descending to ascending aortic conduit and pulmonary artery banding. One month later, the repair was completed with VSD closure and pulmonary artery band removal. A one-stage primary repair of IAA/LVOTO was first reported in 1987 by Yasui and colleagues [2]. Two patients had an 8-mm ascending aorta to descending aorta interposition graft, a Damus-Stansel-Kaye connection of the proximal pulmonary trunk to the small ascending Accepted for publication Feb 8, Presented at the Fifty-eighth Annual Meeting of the Southern Thoracic Surgical Association, San Antonio, TX, Nov 9-12, Address correspondence to Dr Kanter, Pediatric Cardiac Surgery, Emory University School of Medicine, 1405 Clifton Rd, Atlanta, GA 30322; kkanter@emory.edu. aorta, closure of the VSD channeling the left ventricular outflow through the VSD to the pulmonary valve, and construction of a right ventricle-to-pulmonary artery conduit (RV-PA) with a 14-m Hancock valve (Medtronic, Minneapolis, MN). The concepts of the Yasui operation (redirection of left ventricular outflow through the VSD to the pulmonary valve, placement of a conduit from the right ventricle to the pulmonary circulation) were also applied to infants with AA/VSD as a primary repair, initially with a conduit from the pulmonary trunk (the new systemic outflow) [3, 4], and then with a Norwoodtype arch reconstruction [5 7]. In addition to use as a primary neonatal repair, the Yasui operation has also been employed as a staged repair both for IAA/LVOTO [8 12] and for AA/VSD [6, 12]. With this background in mind, we reviewed our experience with 21 infants from 2002 to 2011 with either IAA/LVOTO or AA/VSD for whom the Yasui operation was applied, either as a primary or as a staged repair. Patients and Methods Approval for this retrospective study was obtained from the Emory University School of Medicine Human Investigation Committee and the Institutional Review Board of 2012 by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc doi: /j.athoracsur

2 2000 KANTER ET AL Ann Thorac Surg YASUI OPERATION 2012;93: Abbreviations and Acronyms AA/VSD aortic atresia or severe aortic valve hypoplasia with two wellformed ventricles and a ventricular septal defect BT modified Blalock-Taussig shunt CoA coarctation DSK Damus-Stansel-Kaye anastomosis IAA/LVOTO interrupted aortic arch with a ventricular septal defect and left ventricular outflow tract obstruction LVOTO left ventricular outflow tract obstruction RV-PA right ventricular to pulmonary artery VSD ventricular septal defect Children s Healthcare of Atlanta. The need for patient or family consent was waived. Patient Population From January 2002 to October 2011, 21 neonates presented with either interrupted aortic arch and severe left ventricular outflow tract obstruction (IAA/LVOTO; n 13), aortic atresia or severe aortic valve hypoplasia (aortic annulus 3 mm) with a VSD and 2 well-formed ventricles (AA/VSD; n 7) or interrupted aortic arch with aortic atresia, VSD, and aortopulmonary window (n 1). All had 2 well-formed ventricles. Table 1 shows the patient characteristics at the time of the initial operation. The age at initial operation was similar between the IAA/LVOTO patients ( days) and the AA/VSD patients ( days; p 0.41), as was the weight ( vs kg; p 0.71). Thirteen patients had IAA/LVOTO; 11 with type B IAA, 1 with type A (patient 8 with an aortic valve annulus diameter of 4.1 mm (z score 3.15) and a LVOT diameter of 4.0 mm), and 1 with type C. Eight patients (62%) had an aberrant retroesophageal right subclavian artery (Table 1). Nine of the 13 IAA/LVOTO patients (69%) had a genetic syndrome; 8 with 22q11 microdeletion (DiGeorge syndrome) and 1 with mosaic trisomy 22. The diameter of the aortic annulus measured by echocardiography ranged from 2.1 to 5.1 mm (mean 4.0 mm), with a mean calculated z score of 4.2 (range 2.6 to 8.2). The subaortic diameter ranged from 2.0 to 4.7 mm (mean 3.6 mm). Only 1 of the 13 IAA/LVOTO patients (patient 4) had an aortic valve z score larger than 3.0. During the same time, 44 other neonates had conventional primary biventricular repair of interrupted aortic arch and VSD involving closure of the VSD with or without left ventricular outflow tract resection and repair of the interrupted arch. Seven patients with a VSD and 2 well-formed ventricles had either severe LVOT obstruction defined as an Table 1. Patient Characteristics at the Time of Original Operation Patient Number Age (Days) Weight (kg) Anatomic Group Operative Strategy Syndrome IAA Type Aberrant Right Subclavian IAA/LVOTO Staged DiGeorge Type B Yes IAA/LVOTO Staged DiGeorge Type B Yes IAA/LVOTO Staged DiGeorge Type B No IAA/LVOTO Staged None Type B No IAA/LVOTO Staged DiGeorge Type B Yes IAA/LVOTO Staged None Type B No IAA/LVOTO Staged None Type B No IAA/LVOTO Staged None Type A No IAA/LVOTO Staged DiGeorge Type B Yes IAA/LVOTO Staged DiGeorge Type B Yes IAA/LVOTO Staged DiGeorge Type B Yes IAA/LVOTO Primary Mosaic trisomy 22 Type B Yes IAA/LVOTO Primary DiGeorge Type C Yes Aortic hypoplasia/vsd Staged None Tight CoA Yes Aortic hypoplasia/vsd Staged None Tight CoA Yes Aortic atresia/vsd Staged None Yes Aortic atresia/vsd Primary None No Aortic atresia/vsd Primary None Tight CoA No Aortic atresia/vsd Primary None No Aortic atresia/vsd Primary None No Aortic atresia/iaa/vsd/ aortopulmonary window Staged Cat Eye syndrome Type B Yes CoA coarctation of the aorta; IAA interrupted aortic arch; IAA/LVOTO interrupted aortic arch with left ventricular outflow tract obstruction; VSD ventricular septal defect.

3 Ann Thorac Surg KANTER ET AL 2012;93: YASUI OPERATION 2001 aortic valve annulus less than 3 mm (n 2) or AA (n 5). All had a hypoplastic aortic arch and 3 had a tight coarctation. None had a genetic syndrome (Table 1). The incidence of genetic syndrome was statistically higher in the IAA/LVOTO group compared with the AA/VSD group (69% vs 0%; p 0.05). The final patient (patient 21, Table 1) had aortic atresia, type B interrupted aortic arch and an aortopulmonary window. Genetic evaluation revealed that he had a supernumerary isodicentric chromosome 22 (cat-eye syndrome). Operative Technique We have described the surgical technique for the Yasui operation previously [13]. Patients had either a primary one-stage repair in the neonatal period or a staged repair depending on the characteristics of the patient and surgeon preference. Six patients had a primary repair (4 with AA/VSD and 2 with IAA/ LVOTO); the remainder had a staged repair (Table 2). Primary repair consisted of channeling the VSD to the pulmonary valve, reconstructing the aortic arch with a Norwood-type homograft patch (including primary repair of the interrupted aortic arch, if present), and creation of RV-PA continuity with a valved conduit. Initial palliation for the staged patients involved a Norwood aortic arch reconstruction as with the primary repair. Pulmonary blood flow was provided by a modified Blalock-Taussig (BT) shunt or a RV-PA shunt (Sano shunt). The atrial septum was not opened. The eventual biventricular repair in the staged patients channeled the VSD to the pulmonary valve with enlargement of the VSD (if necessary), creation of RV-PA continuity with a valved conduit, and takedown of the shunt. Statistics Continuous variables were compared by analysis of variance and are presented as the mean standard deviation. Nominal variables were compared by 2 analysis with the Fisher exact test. Life table analysis was done with the Kaplan-Meier survival method with significance determined by log-rank analysis. Significance was defined as a p value less than or equal to Results Early Results Operative details are presented in Table 2. The incidence of a primary Yasui procedure was more common in the AA/VSD patients (57%) than in the IAA/LVOTO patients (15%), although this was not statistically different (p 0.12). There were no early deaths in either group. There were no interstage deaths in the staged patients. In the staged patients, the interval from initial palliation to the eventual Yasui repair ranged from 4.3 to 26.6 months (mean months; median 6.9 months). Patient 1 had an interim bidirectional cavopulmonary anastomosis that was taken down at the time of the repair. Two-staged patients less than 4 months after initial palliation (patients 21 and 11) await biventricular repair. Before April 2003, all staged patients had a modified BT shunt for pulmonary blood flow. Since then, a RV-PA conduit (Sano shunt) has been used in all except the most recent case, patient 21. Due to multiple factors, his operation was delayed until he was 41 days old. Concerned about postoperative pulmonary artery hypertension, he had a 3.5-mm modified BT shunt from his right carotid artery (he had an aberrant retroesophageal right subclavian artery originating from the distal aortic arch). Sixteen days postoperatively, due to hypoxia, he required conversion to a RV-PA conduit with takedown of his BT shunt. He has done well since then and awaits his reparative operation. With the biventricular repair, whether primary or staged, all patients had a Yasui-type operation except patient 4 (the only IAA/LVOTO patient with an aortic annulus z score greater than 3.0; his z score was 2.6). He had takedown of his Norwood proximal aorta-topulmonary connection, resection of LVOTO, VSD closure, and reestablishment of continuity between his native pulmonary valve and his branch pulmonary arteries. He is the only patient in this series to require reoperation for recurrent LVOTO (33 months after his biventricular repair). The remaining 18 repaired patients had routing of the left ventricular outflow through the VSD (enlarged in 7 of the 13 staged repairs and none of the primary repairs; Table 2). No patient required a pacemaker. One child (patient 9) had neoaortic valve replacement (the native pulmonary valve) at the time of biventricular repair due to significant neoaortic insufficiency not amenable to valve repair. Right ventricular to pulmonary artery continuity was established with a valved homograft or a valved bovine jugular vein conduit. The RV-PA conduits used in the primary repair patients were significantly smaller than in the staged patients ( mm vs mm; p ). Patient 2, who had his initial palliation in 2002, is the only operative mortality after biventricular repair; he died 2 months postoperatively of multisystem organ failure. Follow-Up On average, follow-up of years (range 5 weeks to 9.5 years), there have been 3 late deaths; 2 in the staged group of infection and respiratory failure 3.4 months (patient 9) and 8.1 months (patient 5) after biventricular repair and 1 in the primary repair group of cerebral hemorrhage 3.1 years postoperatively (patient 12). All 4 deaths were in patients with a genetic syndrome (Di- George syndrome in 3, trisomy 22 in 1) and with IAA/ LVOTO. Overall freedom from death after the initial operation was 100% at 1 year and 75% at 5 years (Fig 1). There were no statistically significant differences in survival between the patients with a primary Yasui procedure and those with a staged repair. Five patients had balloon dilatation of a recurrent arch obstruction or coarctation in the catheterization laboratory 3.1 to 7.9 months (mean 5.3 months) postoperatively.

4 Table 2. Operative Details Patient Number Operative Repair Strategy Anatomic Group Sano/Blalock-Taussig Shunt Interval from Initial Operation to Repair (months) RV-PA Valved Conduit RV-PA Conduit Size (mm) 17 Primary Aortic atresia/vsd 0.0 Aortic homograft 9 No 18 Primary Aortic atresia/vsd 0.0 Aortic homograft 10 No 19 Primary Aortic atresia/vsd 0.0 Aortic homograft 9 No 20 Primary Aortic atresia/vsd 0.0 Pulmonary homograft 10 No 12 Primary IAA/LVOTO 0.0 Aortic homograft 10 No 13 Primary IAA/LVOTO 0.0 Aortic homograft 10 No Mean SD for primary repair patients a 16 Staged Aortic atresia/vsd Sano 4.4 Bovine jugular vein 12 Yes 14 Staged Aortic hypoplasia/vsd Sano 12.6 Aortic homograft 16 Yes 15 Staged Aortic hypoplasia/vsd Sano 5.2 Bovine jugular vein 14 Yes 1 b Staged IAA/LVOTO BT 26.6 b Pulmonary homograft 16 No 2 Staged IAA/LVOTO BT 10.8 Aortic homograft 14 No 3 Staged IAA/LVOTO BT 6.4 Aortic homograft 15 No 4 c Staged IAA/LVOTO BT 6.4 c c c 5 Staged IAA/LVOTO Sano 10.1 Pulmonary homograft 14 Yes 6 Staged IAA/LVOTO Sano 4.3 Bovine jugular vein 14 Yes 7 Staged IAA/LVOTO Sano 6.9 Pulmonary homograft 14 Yes 8 Staged IAA/LVOTO Sano 17.6 Pulmonary homograft 13 No 9 Staged IAA/LVOTO Sano 22.6 Pulmonary homograft 11 No 10 Staged IAA/LVOTO Sano 6.8 Aortic homograft 13 Yes 11 Staged IAA/LVOTO Sano Awaiting repair 21 Staged Aortic atresia/iaa/vsd/ aortopulmonary window BT then Sano Awaiting repair Mean SD for staged repair patients (median 6.9 months) a VSD Enlarged a p primary repair versus staged repair. b Patient 1 had an interim bidirectional cavopulmonary anastomosis between initial palliation and Yasui repair. c Patient 4 had DSK takedown, LVOTO resection, and VSD closure at time of repair. BT modified Blalock-Taussig shunt; DSK Damus-Stansel-Kaye anastomosis; IAA interrupted aortic arch; IAA/LVOTO interrupted aortic arch with left ventricular outflow tract obstruction; LVOTO left ventricular outflow tract obstruction; RV-PA right ventricle to pulmonary artery; VSD ventricular septal defect KANTER ET AL Ann Thorac Surg YASUI OPERATION 2012;93:

5 Ann Thorac Surg KANTER ET AL 2012;93: YASUI OPERATION 2003 Fig 1. Kaplan-Meier actuarial freedom from death after the initial operation. There were no statistically significant differences in survival between the patients who had a primary ( ) Yasui procedure and those who had a staged ( ) repair. Reoperations After biventricular repair, 8 of the 18 survivors (44%) required reoperation 6 months to 5 years after the biventricular repair (mean 3.2 years). Except for the abovementioned patient 4 with recurrent LVOTO after a standard biventricular repair, no patient developed LVOTO. Patient 1 required closure of a residual VSD and tricuspid valve repair 6 months after a staged biventricular repair. The remaining 6 reoperations were for replacement of stenosed RV-PA conduits; one of these reoperations included an associated neoaortic valve replacement (patient 12) and another had surgical correction of residual arch obstruction at the time of the conduit change (patient 16). Actuarial freedom from reoperation or death after biventricular repair was 78% at 1 year and 14% at 5 years. There were no statistically significant differences in actuarial freedom from reoperation or death between the patients who had a primary Yasui procedure and those who had a staged repair, despite the fact that the primary repair patients had significantly smaller RV-PA conduits placed at the time of the biventricular repair (Fig 2). with IAA/LVOTO or AA/VSD. Actuarial 5-year survival in the staged group was 89% compared with 73% for the primary Yasui group, but this did not achieve statistical significance. Although their data supported either a primary or staged approach, the authors recommended primary biventricular repair. A more recent series from Children s Hospital, Boston [20], examined the results with a primary Yasui repair in 17 neonates with AA/VSD (n 7) or IAA/LVOTO (n 5). Despite 3 early deaths (17.6% 30-day mortality), the 10-year actuarial survival was 82%. Freedom from reoperation at 3 years was less than 40%. A similar study of a primary Yasui operative approach in 21 neonates from Philadelphia [21] reported no operative mortality but intraventricular baffle leak in 11 patients, 4 of whom required reoperation. On median follow-up of 34.2 months, there was 1 late death. Even with their relatively short follow-up, half of the survivors required at least 1 reoperation. Another report of primary Yasui repair in 16 infants from Birmingham, England [22], reported a 19% early mortality with a 60% 1-year actuarial survival and 5-year freedom from reoperation of 20%. This current report demonstrates that the Yasui operation can be applied safely to infants with IAA/LVOTO or AA/VSD, either as a primary or as a staged approach. For the initial operation, there were no operative deaths in either group with a 1-year actuarial survival after the initial operation of 100% (Fig 1). There was 1 postoperative death at the time of biventricular repair in a staged patient with a 5-year actuarial survival of 75%. No patient required a pacemaker even though 7 had VSD enlargement at the time of biventricular repair (all in the staged group; Table 2). The only patient who developed recurrent LVOTO was the only patient in this series who had VSD closure with preservation of left ventricular to aortic continuity without a Rastelli-type approach (patient 4). No other patient developed LVOTO either in the primarily repaired group or the staged Comment The Yasui operation is a useful approach for the management of patients with IAA/LVOTO or with AA/VSD. Since its introduction in 1987 [2], several case reports or small series have described the utility of this operation [3 5, 7, 9 11, 14 19]. Early reports from Philadelphia describing children with IAA/LVOTO recommended a staged repair over a primary Yasui operation because of a lower overall mortality [6, 8]. Ohye and colleagues from the University of Michigan [12] reviewed their experience with primary and staged Yasui operations in 20 infants Fig 2. Kaplan-Meier actuarial freedom from reoperation or death after biventricular repair. There were no statistically significant differences in survival between the patients who had a primary ( ) Yasui procedure and those who had a staged ( ) repair.

6 2004 KANTER ET AL Ann Thorac Surg YASUI OPERATION 2012;93: group. This freedom from LVOTO has been reported by others with the Yasui operation [20 24]. Our choice of a primary or a staged approach was based on the clinical status of the patient and the discretion of the surgeon. Although more patients with AA/ VSD had a primary repair than those with IAA/LVOTO, this was not statistically different. Mortality rates and the need for reoperation were comparable between the 2 approaches (Figs 1 and 2) as has been demonstrated by others [12]. Despite concerns raised by others about interstage mortality in patients initially palliated with a Norwood operation and staged to a Yasui repair [20], there were no interstage deaths in this series as we [10] and others [11] have previously reported. For patients with IAA/LVOTO, predicting which patients have a left ventricular outflow tract too small for an anatomic repair (closure of the VSD with maintenance of left ventricular to aortic continuity with or without subaortic resection) and will thus need a Yasui approach can be difficult. The 13 patients with IAA/LVOTO in this series managed with a Yasui approach represent 23% of the 57 neonates presenting to our institution between 2002 and 2011 with interrupted aortic arch and VSD. This is similar to the experience from Michigan; 15 of 50 neonates (30%) presenting with interrupted aortic arch had such severe left ventricular outflow tract obstruction that a Yasui approach was used [25]. Only one of our IAA/LVOTO patients had an aortic annulus z score greater than 3.0. In general, we apply the Yasui approach to neonates in whom the smallest diameter of the left ventricular outflow tract (either at the valve or the subaortic region) is less than 4.0 mm. For those with a minimum value between 4.0 and 4.5 mm, we would consider either a conventional repair or a Yasui, and for those patients with a value greater than 4.5 mm, we would tend to perform a standard biventricular repair. Although we have previously described the use of the Ross-Konno procedure in patients with IAA/LVOTO after initial Norwood palliation [10], we now reserve the Ross-Konno for patients who have had a more standard anatomic interrupted aortic arch repair and develop recurrent LVOTO. The high reoperation rate in this series (Fig 2) was mostly due to the need for RV-PA conduit replacement. This is not dissimilar to other institutions experience [20 23] and is not unexpected when one uses valved RV-PA conduits in small babies. It is important to note that all reoperations in this series were performed safely without mortality. All 4 deaths in this series were in patients with genetic syndromes (DiGeorge syndrome in 3, trisomy 22 in 1). The association of genetic syndrome and poorer longterm outcome previously has been reported in patients with pulmonary atresia and VSD [26] and in patients undergoing a Norwood procedure [27]. Interestingly, in this series there were no early or late deaths in the AA/VSD patients. This lower mortality in patients with AA/VSD compared with those with IAA/LVOTO previously has been noted in a multiinstitutional Congenital Heart Surgeons Society study [24] as well as a paper from Saudi Arabia [23]. Perhaps this favorable effect is due to a lower incidence of genetic syndromes in the AA/VSD patients compared with the IAA/LVOTO patients (none of our patients with AA/VSD had a genetic syndrome; Table 1). In summary, the Yasui operation is effective for children with IAA/LVOTO and with AA/VSD. Primary and staged repair have comparable results. Reoperation after biventricular repair seems inevitable, mostly for conduit replacement. Genetic factors may affect long-term survival. References 1. Norwood WI, Stellin GJ. Aortic atresia with interrupted aortic arch: reparative operation. J Thorac Cardiovasc Surg 1981;81: 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: Austin EH, Jonas RA, Mayer JE, Castaneda AR. Aortic atresia with normal left ventricle. Single-stage repair in the neonate. J Thorac Cardiovasc Surg 1989;97: Serraf A, Bruniaux J, Lebidois J, Lacour-Gayet F, Kachaner J, Planche C. Aortic atresia with normal left ventricle. Ann Thorac Surg 1991;51: Bogers AJ, Sreeram N, Hess J, Sutherland GR, Quaegebeur JM. Aortic atresia with normal left ventricle: one-stage repair in early infancy. Ann Thorac Surg 1991;51: Rychik J, Murdison KA, Chin AJ, Norwood WI. Surgical management of severe aortic outflow obstruction in lesions other than the hypoplastic left heart syndrome: use of a pulmonary artery to aorta anastomosis. J Am Coll Cardiol 1991;18: Francois K, Dollery C, Elliott MJ. Aortic atresia with ventricular septal defect and normal left ventricle: one-stage correction in the neonate. Ann Thorac Surg 1994;58: Jacobs ML, Chin AJ, Rychik J, Steven JM, Nicolson SC, Norwood WI. Interrupted aortic arch. Impact of subaortic stenosis on management and outcome. Circulation 1995;92(9 suppl):ii Kanter KR, Miller BE, Cuadrado AG, Vincent RN. Successful application of the Norwood procedure for infants without hypoplastic left heart syndrome. Ann Thorac Surg 1995;59: Erez E, Tam VK, Kanter KR, Fyfe DA. Successful biventricular repair after initial Norwood operation for interrupted aortic arch with severe left ventricular outflow tract obstruction. Ann Thorac Surg 2001;71: Pearl JM, Cripe LW, Manning PB. Biventricular repair after Norwood palliation. Ann Thorac Surg 2003;75: Ohye RG, Kagisaki K, Lee LA, Mosca RS, Goldberg CS, Bove EL. Biventricular repair for aortic atresia or hypoplasia and ventricular septal defect. J Thorac Cardiovasc Surg 1999;118: Kanter KR. The Yasui operation. Operat Tech Thorac Cardiovasc Surg 2010;15: Black MD, Smallhorn JF, Freedom RM. Aortic atresia with a ventricular septal defect: modified single-stage neonatal biventricular repair. Ann Thorac Surg 1999;67: Gandhi SK, Siewers RD, Pigula FA. Complete surgical repair of aortic atresia with a normal left ventricle. Ann Thorac Surg 2000;70: Nurozler F, Bradley SM. Considerations in biventricular repair after the Norwood procedure. Eur J Cardiothorac Surg 2000;18: Yamagishi M, Fujiwara K, Yamada Y, Shuntoh K, Kitamura N. A new surgical technique for one-stage repair of interrupted aortic arch with valvular aortic stenosis. J Thorac Cardiovasc Surg 2001;122:392 3.

7 Ann Thorac Surg KANTER ET AL 2012;93: YASUI OPERATION Takabayashi S, Kado H, Shiokawa Y, Fukae K, Nakano T. Long-term outcome of left ventricular outflow tract after biventricular repair using Damus-Kaye-Stansel anastomosis for interrupted aortic arch and severe aortic stenosis. J Thorac Cardiovasc Surg 2005;130: Uemura H, Adachi I, Kagisaki K, Shikata F, Yagihara T. Intermediate results after a modified Yasui procedure with the Lecompte maneuver. Ann Thorac Surg 2007;84: Nathan M, Rimmer D, del Nido PJ, et al. Aortic atresia or severe left ventricular outflow tract obstruction with ventricular septal defect: results of primary biventricular repair in neonates. Ann Thorac Surg 2006;82: Gruber PJ, Fuller S, Cleaver KM, et al. Early results of single-stage biventricular repair of severe aortic hypoplasia or atresia with ventricular septal defect and normal left ventricle. J Thorac Cardiovasc Surg 2006;132: Moorthy PS, McGuirk SP, Jones TJ, Brawn WJ, Barron DJ. Damus-Rastelli procedure for biventricular repair of aortic atresia and hypoplasia. Ann Thorac Surg 2007;84: DISCUSSION DR EDWARD L. BOVE (Ann Arbor, MI): Dr Kanter, that is an excellent series and I really enjoyed your paper. You have outstanding results in a particularly high-risk group of neonates. Your paper highlights the very high incidence of extracardiac abnormalities, particularly chromosomal defects, which were a risk factor for outcome, a finding that has been replicated in many, many other studies looking at all types of congenital heart defects. Our experience at the University of Michigan is very similar to yours, even including the fact that we had a mixture of singlestage and two-stage repairs. Early in our own experience, which we reported some years ago, we selected the highest risk patients, low birth weight, prematurity, chromosomal anomalies, et cetera, for single-stage repairs, believing that these patients were likely to fare poorly with a shunt-dependent circulation. When we reviewed our data we found, as you did, that there was no difference, which significantly influenced our own group into eliminating two-stage repairs and going to a one-stage repair essentially for all of these patients. I have a few questions for you and the excellent data that you showed. First, how did you determine when the ventricular septal defect needed to be enlarged or did you do that routinely for those patients who had a Yasui repair? Second, at least it seems to me that one of the major difficulties with this operation from a technical point of view is fashioning and inserting a conduit in some of these very tiny babies when the conduit sits so anteriorly in the chest. Do you have any tricks about that? And might you comment on those that feel that a posterior translocation of the pulmonary root, like a Ross procedure, is a better approach, because it allows a more orthotopic position for the right ventricular outflow tract conduit? And along those same lines, would you then consider perhaps not placing a conduit or using a transannular patch to try to decrease the reoperation rate? And then last, what will you do now looking at your own data? How will you determine when the next baby comes along whether you perform a single versus two-stage approach? What criteria do you and your colleagues use? Thank you. I enjoyed the paper immensely. DR KANTER: Thank you, Ed. Your group clearly is a pioneer in this operation and have published two very important papers on it. 23. Alsoufi B, Al-Halees Z, Manlhiot C, et al. Intermediate results following complex biventricular repair of left ventricular outflow tract obstruction in neonates and infants. Eur J Cardiothorac Surg 2010;38: Hickey EJ, Yeh T Jr, Jacobs JP, et al. Ross and Yasui operations for complex biventricular repair in infants with critical left ventricular outflow tract obstruction. Eur J Cardiothorac Surg 2010;37: Suzuki T, Ohye RG, Devaney EJ, et al. Selective management of the left ventricular outflow tract for repair of interrupted aortic arch with ventricular septal defect: management of left ventricular outflow tract obstruction. J Thorac Cardiovasc Surg 2006;131: Mahle WT, Crisalli J, Coleman K, et al. Deletion of chromosome 22q11.2 and outcome in patients with pulmonary atresia and ventricular septal defect. Ann Thorac Surg 2003;76: Gaynor JW, Mahle WT, Cohen MI, et al. Risk factors for mortality after the Norwood procedure. Eur J Cardiothorac Surg 2002;22:82 9. We evaluate the VSD [ventricular septal defect] in the operating room, and if it looks like it is going to be restrictive, we will enlarge it. Certainly if it is less than the diameter of the pulmonary valve, which is going to be the neoaortic valve, then we will enlarge the VSD. In the infants undergoing a primary repair, none of them required ventricular septal defect enlargement, because all had large VSDs. Therefore, it is a subjective approach. We don t do it routinely; only about half the patients in the staged group had a VSD enlargement. The conduit is extraanatomic and is right behind the sternum, which can be challenging on sternal reentry. We do make an effort to bring it as far lateral as possible. Certainly for the staged patients who will have an RV [right ventricle] to PA [pulmonary artery] conduit for the original operation, we construct the Sano as a marking point and bring it well over to try to put it more in the left chest so it is away from the sternum. However, we are not always able to accomplish that. The use of the Ross procedure is intriguing. For our patients with interrupted aortic arch, we have saved the Ross procedure for the reoperation for the patients who have a bicuspid aortic valve or a marginal left ventricular outflow tract. About threequarters of our interrupted arch patients had a more standard repair. A fair number of them have come back, I would say at least 10 or 20%, requiring aortic valve replacement, and then we will do a Ross-Konno procedure. I am not sure how you could do a Ross-Konno procedure in a patient with aortic atresia, especially with an aorta of 1 mm, but certainly that s a possibility for a patient with an interrupted aortic arch and left ventricular outflow tract obstruction. The thought of doing a primary repair with a Ross-Konno, a Norwood arch reconstruction, interrupted arch repair just seems a little daunting to me. So the first time we do it, we are going to ask you to come down and scrub on the case. Certainly a transannular patch for the right ventricular outflow tract is an attractive idea, and there are people who have discussed the use of a transannular patch for the very reason of avoiding conduit reoperation. The pulmonary vascular resistance in these infants is high, and we are worried about early postoperative right ventricular failure and have not done so. However, everyone s experience with neonatal correction of pulmonary atresia and VSD we know that we can get by without a valve, but that is different physiology early on. We have just not been bold enough to do it, because these operations, as you

8 2006 KANTER ET AL Ann Thorac Surg YASUI OPERATION 2012;93: well know, are very long with long cross-clamp times. So we are afraid not to have a valve in that circuit. What would we do now with a patient who came along? I think those patients with aortic atresia we would tend to do a primary repair. Those with interrupted arch and left ventricular outflow tract obstruction, we might do a primary repair, but we still might stage them. And the reason I say that is the one tantalizing patient whose left ventricular outflow tract grew, so that eventually we were able to go back and do an anatomic repair, a VSD closure using his native aortic valve for left ventricular outflow. That was the only patient in the series whose z value was greater than 3. His was 2.6. DR JOHN EDMUND MAYER, JR (Boston, MA): Just to follow up on that last point, what is your cutoff as you are thinking about this for what is an adequate left ventricular outflow tract that could be done as a two-ventricle repair just closing the VSD and repairing the arch, as opposed to this sort of approach where you think the left ventricular outflow tract is basically not going to sustain systemic cardiac output? DR KANTER: Well, we don t have a hard number and it is certainly subjective, and sometimes we go to the operating room, especially with the interrupted arch with left ventricular outflow tract obstruction, with a plan to do an anatomic repair and then bail out to the Yasui operation. Clearly, anyone with an aortic valve annulus with a z score larger than 2 we would not use this approach, and, in fact, all but one in this study had a z score less than 3. So I think that would be a number that we would hang our hat on. Looking at millimeters, for an average size baby between 2.5 and 3.5 kg, we would say if the aortic annulus is 5 mm, we would never consider this operation, between 4.5 to 5 mm, we would wonder about it, and under 4.5 and certainly under 4 mm, we would do a Yasui procedure. DR PETER B. MANNING (Cincinnati, OH): You mentioned that early in your series you primarily did B-T [Blalogh-Taussig] shunts, and later performed Sanos. Does this reflect a shift in your institutional bias toward which shunt you perform for all Norwoods, including those with hypoplastic left heart syndrome, or is it unique to this population where you know these kids are going to have an RV to PA conduit at some time? DR KANTER: It had the desirability of having your ventriculotomy already made and the marking point, so that was one reason, but at the same time we have gravitated towards the Sano shunt in our institution as the preferred shunt for Norwood procedures. We don t do it exclusively, but we do it 90% of the time. DR BOVE: I just want to clarify one thing, Kirk. Before I get the invitation to come down and scrub on the case, I wouldn t do it any differently than you do, and I am not an advocate of the Ross. I just wanted to hear your thoughts. But I d love the invitation. DR CHRISTOPHER BAIRD (Boston, MA): I just wanted to follow up on the comments to Dr Mayer s question. One of the things we have been doing more recently in Boston is looking at the left ventricular outflow tract with 3D echo. It gives us a little bit better assessment, because a lot of these outflow tracts appear to be oval instead of a complete circle. And so making our dimensions in one plane, it fooled us sometimes. So we have critically been trying to assess that ahead of time, and it has given us a pretty good sense going into the operation that these oval type outflow tracts are still amenable to primary repair. DR KANTER: There is actually a paper I think by Adey and colleagues that did not use 3-dimensional but developed a cross-sectional area of the left ventricular outflow tract and found that some value, which I don t remember, centimeter squared was predictive of left ventricular outflow tract obstruction as opposed to those who would definitely need this approach. So I think that that in the future will help us answer John Mayer s question about which are the patients who should have this approach as opposed to the more standard anatomic repair of interrupted arch with VSD.