Aortic Arch Advancement: The Optimal One-Stage Approach for Surgical Management of Neonatal Coarctation With Arch Hypoplasia

Aortic Arch Advancement: The Optimal One-Stage Approach for Surgical Management of Neonatal Coarctation With Arch Hypoplasia Mohamed-Adel Elgamal, MD, E. Dean McKenzie, MD, and Charles D. Fraser, Jr, MD Congenital Heart Surgery Service, Texas Children s Hospital, and Division of Congenital Heart Surgery, Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, Texas Background. The optimal surgical treatment for neonatal coarctation with aortic arch hypoplasia (NCoAo/AAH) is controversial. Important long-term concerns include arch growth. We report our results obtained with a one-stage radical approach of coarctectomy and aortic arch advancement for NCoAo/AAH. Methods. From June 1995 to December 2000, 65 newborns with NcoAo/AAH underwent coarctectomy and aortic arch advancement via a median sternotomy under deep hypothermic circulatory arrest. Patients were classified by diagnosis: group 1, isolated NCoAo/AAH (n 13); group 2, NCoAo/AAH with ventricular septal defect (n 20); and group 3, NCoAo/AAH with complex cardiac lesions (n 32). Results. The study population included 36 boys and 29 girls. Mean age was 13 1.7 days (range 1 to 43 days). Mild to moderate left ventricular outflow tract obstruction was present in 15 patients. Mean body weight was 3.4 0.1 kg (range 1.6 to 5 kg). Eight babies were premature. The mean Z value for the aortic arch was 4 0.3 (range 2 to 4.5) and for the isthmus 4.5 0.2 (range 3 to 7). Mean deep hypothermic circulatory arrest time was 28 2 minutes (range 14 to 60 minutes). Mean intensive care unit stay was 6 1 days (range 2 to 30 days). There were three early deaths (all in groups 2 and 3) and two late deaths (in group 3) (5-year actuarial survival, 91% 7.9%). There was one recurrence (5-year actuarial freedom from recurrence, 98% 4%). Peak Doppler velocity across the arch in the remaining patients was 1 0.1 m/s (range 0 to 2.2 m/s). Conclusions. Coarctectomy and aortic arch advancement is the optimal surgical method for management of NCoAo/AAH. It has low operative morbidity and mortality and a very low incidence of recoarctation or arch obstruction. (Ann Thorac Surg 2002;73:1267 73) 2002 by The Society of Thoracic Surgeons Since Mustard and associates performed the first successful repair of neonatal coarctation in 1953, the results of surgical management of this congenital anomaly have significantly improved [1]. Complex neonatal coarctation associated with severe arch hypoplasia, however, remains a medical and surgical challenge [2]. Furthermore, the relative benefits of staged versus one-stage repair of this lesion when associated with intracardiac defects remain debatable [3, 4]. Although there is agreement that successful repair of neonatal coarctation should address complete relief of arch obstruction, there is still no consensus about the ideal surgical technique [2, 5 8]. Recent refinements in the management of neonates with complex congenital cardiac defects and the increasing use of two-dimensional (2-D) echo-doppler evaluation instead of cardiac catheterization have allowed onestage repair of complex congenital cardiac and aortic arch Presented at the Thirty-seventh Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 29 31, 2001. Address reprint requests to Dr Fraser, Congenital Heart Surgery Service, Texas Children s Hospital/Baylor College of Medicine, 6621 Fannin St, Mail Code: WT 19345-H, Houston, TX 77030-2399; e-mail: charlesf@ bcm.tmc.edu. anomalies to be done completely, safely, and effectively even in premature and low-birth-weight newborns [3, 9]. Since June 1995, we have utilized a one-stage surgical strategy for repair in all neonates presenting at our institution with neonatal coarctation with aortic arch hypoplasia (NCoAo/AAH). The strategy involves aortic arch advancement via a median sternotomy during a brief period of deep hypothermic circulatory arrest (DHCA). We report here our experience with this strategy, which includes resection of the coarctation and aortic arch reconstruction in a way that completely eliminates the coarctation and aortic arch hypoplasia. Material and Methods Patient Population From June 1995 to December 2000, 65 consecutive neonates (&lteq; 45 days old) at our institution underwent surgical repair of NCoAo/AAH by means of a one-stage radical approach involving coarctectomy and aortic arch advancement via median sternotomy during brief DHCA. All patients underwent definitive repair of their associated intracardiac defects, with the exception of 2002 by The Society of Thoracic Surgeons 0003-4975/02/$22.00 Published by Elsevier Science Inc PII S0003-4975(01)03622-0

1268 ELGAMAL ET AL Ann Thorac Surg AORTIC ARCH ADVANCEMENT 2002;73:1267 73 Table 1. Diagnosis of 65 Patients With Neonatal Coarctation With Aortic Arch Hypoplasia Group Description n Group 1 Coarctation with aortic arch hypoplasia 13 Group 2 Coarctation with aortic arch hypoplasia/ 20 ventricular septal defect Group 3 Coarctation with aortic arch hypoplasia/ 32 complex cardiac defect Double-outlet right ventricle (n 7) Complete atrioventricular septal defect (n 9) d-transposition of great arteries (n 5) Shone s complex (n 4) Single ventricle (n 7) those expected to undergo future univentricular palliation. Preoperative Management Prostaglandin E1 (PGE1) infusion was required to maintain ductal patency in 39 patients. All patients exhibited significant congestive heart failure. Fifteen were critically ill and required intensive preoperative resuscitation for hypotension, renal failure, and acidosis. The diagnosis was suspected or made clinically and then confirmed by complete transthoracic 2-D echo-doppler evaluation. Cardiac catheterization was performed in 9 patients to better define the arch anatomy. There was excellent correlation between echo-based diagnosis and operative findings. The Z value of the segment of the transverse arch between the innominate and the left common carotid arteries was calculated; a value of 2 or more was considered a criterion for the diagnosis of arch hypoplasia. We divided the patients by preoperative diagnosis into three groups: group 1, isolated NCoAo/AAH (n 13); group 2, NCoAo/AAH with ventricular septal defect (VSD) (n 20); and group 3, NCoAo/AAH with complex cardiac lesions (n 32). Table 1 summarizes the cardiac diagnosis of the three groups. A bicuspid aortic valve was present in 21 patients. Mild to moderate left ventricular outflow tract obstruction (LVOTO) was present in 15 patients. Extracardiac congenital anomalies were present in 15 patients, as detailed in Table 2. Surgical Technique In all patients, the defect was repaired via a median sternotomy using cardiopulmonary bypass, core cooling to 18 C, and DHCA. Dissection of the ascending aorta and as much as possible of the aortic arch and its vessels was done before cannulation. Snares were also passed around the arch vessels in preparation for DHCA. Single or double arterial (ascending aorta and duct) cannulation was performed as necessary to ensure adequate perfusion distal to the coarctation. When possible, bicaval venous cannulation was utilized to minimize the duration of DHCA. Patients were cooled to 18 C more than 25 to 30 minutes at minimum flow rates of 150 ml/kg/min. During cooling, the upper descending aorta was mobilized as far distally as possible. Radial and femoral (or umbilical-arterial) pressures were continuously measured to ensure adequate flow distribution. The duct was divided and its tissue was trimmed from the extensively mobilized descending aorta. The upper aortic intercostal arteries were rarely divided. A counterincision was then made on the undersurface of the proximal arch opposite to the origin of the innominate artery and extending into the most distal part of the ascending aorta. An end-toside anastomosis was carried out between the descending aorta and the distal part of the ascending aorta. Concomitant repair of associated cardiac defects was then done either during the same or additional periods of DHCA or during cardiopulmonary bypass (CBP), depending on individual case anatomy. Follow-up All hospital survivors were followed up regularly at the discretion of their pediatric cardiologists. Follow-up was complete for all patients. The quality of coarctation and arch repair was assessed by history and clinical examination, including pulse and blood pressure measurement in the four limbs, cardiac examination, and transthoracic 2-D echo-doppler examination. Follow-up cardiac catheterization was performed in 11 patients: 10 as a part of evaluation for subsequent univentricular palliation and one for evaluation of a recurrent coarctation. Follow-up data were recorded and retrieved for analysis. Statistical Analysis Mortality (early and late), actuarial survival rates, and freedom from recurrence of coarctation and reoperation were analyzed by Kaplan-Meier curves expressed with 95% confidence limits (CL). Results Patient Characteristics The study population included 36 boys and 29 girls. Mean age was 13 1.7 days (range 1 to 43 days). Mean body weight was 3.4 0.1 kg (range 1.6 to 5 kg). Eight patients were born premature. Operative Data Mean aortic cross-clamp time was 79 7 minutes (range 20 to 192 minutes). Mean DHCA time was 28 2 minutes Table 2. Associated Extracardiac Lesions Lesion Down s syndrome 5 DiGeorge s syndrome 3 Turner s syndrome 2 Malrotation 1 Beal s syndrome 1 Situs inversus 1 Pelvic hematoma 1 Right inguinal hernia 1 n

Ann Thorac Surg ELGAMAL ET AL 2002;73:1267 73 AORTIC ARCH ADVANCEMENT 1269 Fig 1. Actuarial hospital survival (Kaplan-Meier) for 65 newborns (95% 5% confidence limits [CL]). The solid line is the calculated Fig 2. Actuarial late survival (Kaplan-Meier) for 65 newborns (96% 6% confidence limits [CL]). The solid line is the calculated (range 14 to 60 minutes). Mean intensive care unit stay was 6 1 days (range 2 to 30 days). The mean Z value for the proximal aortic arch was 3 0.2 (range 2 to 3.5), for the distal arch 4 0.3 (range 3 to 4.5), and for the isthmus 4.5 0.2 (range 3 to 7). Early Mortality There were three early deaths (defined as death within the first 30 postoperative days or before hospital discharge). One occurred in group 2 and two in group 3. The 5-year actuarial freedom from early mortality was 95% 5% (95% CL) (Fig 1). All three early deaths were cardiac related. One patient had a ventricular septal defect and presented in a state of severe congestive heart failure at 40 days of age. He developed pulmonary hypertension after surgery and died. A second patient had an unbalanced complete atrioventricular septal defect with left ventricular hypoplasia and subaortic stenosis. He could not be weaned off a ventilator, and he was taken to the catheterization laboratory for evaluation of the degree of LVOTO. During the procedure, he developed ventricular fibrillation and died. The third patient had a complete atrioventricular septal defect. After surgery, he had persistent, severe mitral regurgitation with pulmonary hypertension and eventually died of right-sided heart failure. Late Mortality There were two late deaths (defined as death after the first 30 postoperative days or after hospital discharge), and both occurred in group 3. The 5-year actuarial freedom from late mortality was 96% 6% (95% CL) (Fig 2). The cause of death was LVOTO in 1 patient and unknown in the other. This gave an overall 5-year actuarial survival rate of 91% 7.5% (95% CL) (Fig 3). Morbidity Early complications occurred in 7 patients: pulmonary hypertension in 4, seizures in 1, transient complete heart block in 1, left chylothorax in 1, and severe hypotension and bradycardia necessitating chest compression in 2. Most of these complications occurred in group 3. All 7 patients were treated without any residua or sequelae. Late complications occurred in 2 patients in group 3. One of them developed severe aortic stenosis and underwent the Ross procedure 18 months after his first surgery; the other developed moderately severe subaortic stenosis and is currently being evaluated carefully for the possibility of surgical intervention. Recurrent Coarctation A significant recurrence occurred in 1 patient who had Shone s complex and had undergone resection of a stenosed segment of the aortic arch and primary end-toend anastomosis of the transverse arch. The recurrence, diagnosed by transthoracic 2-D echo-doppler evaluation (peak echo-doppler velocity 4 m/s; catheterization gradient, 55 mm Hg), was due to incomplete removal of the ductal tissue in the first surgery. As shown in Figure 4, the 5-year actuarial freedom from recurrence was 98% 4% (95% CL). The peak velocity acceleration across the aortic arch in the remaining patients, as assessed by transthoracic 2-D echo-doppler evaluation, ranged from 0 to 2.2 m/s (mean 1 0.1 m/s). Comment The incidence of arch hypoplasia in neonates presenting with coarctation varies from 65% [10] to 81% [11]. Arch hypoplasia is more common in cases of neonatal coarctation associated with intracardiac defects, as supported Fig 3. Actuarial overall survival (Kaplan-Meier) for 65 newborns (91% 7.5% confidence limits [CL]). The solid line is the calculated

1270 ELGAMAL ET AL Ann Thorac Surg AORTIC ARCH ADVANCEMENT 2002;73:1267 73 Fig 4. Actuarial freedom from recurrence of coarctation in 65 newborns (98% 4% confidence limits [CL]). The solid line is the calculated by the hemodynamic underdevelopment theory for the pathogenesis of aortic coarctation [12]. Several pathologic studies on infantile coarctation have highlighted the important association with tubular hypoplasia of the transverse arch [13, 14]. It is highly possible that the introduction of PGE1 therapy has modified the anatomic types of neonatal coarctation that are currently seen, by generating an increasing number of patients with coarctation and severe hypoplasia [15]. It may be assumed that most neonates with NCoAo/AAH now seen in the post- PGE1 era by pediatric cardiologists and treated by pediatric cardiac surgeons are those who were seen in the pre-pge1 era by pathologists. Usually, these neonates present in a state of low cardiac output during the first few days of life. This is especially true when complex intracardiac anomalies are present. Our series includes only patients with NCoAo/AAH. Since we started utilizing our one-stage approach, we have seen an annual increase in the number of patients treated with it. This may be due to the increased experience and our lower threshold for performing this new approach, together with improved early diagnosis and resuscitation, allowing increased referral to our center. Yet, even though the surgical management of NCoAo/ AAH has significantly improved in terms of morbidity, mortality, and recoarctation, there is still no consensus about the optimal surgical strategy. Issues under debate are [1] a precise definition of arch hypoplasia [2], onestage versus staged repair of NCoAo/AAH and the optimal surgical technique for doing so, and [3] subsequent growth of the aortic arch. Definition of Aortic Arch Hypoplasia Formulating a precise definition of aortic arch hypoplasia and of what is small and what is hypoplastic is very difficult [16]. Some degree of aortic arch hypoplasia is present in many neonates presenting with coarctation [6, 12, 13, 16]. It is of paramount importance to determine if such hypoplasia is severe enough to create flow obstruction. We question whether any neonate presenting with coarctation and duct-dependent perfusion has an adequate arch. Several recent pathologic [14] and surgical [2, 5, 6, 7, 16] reports have emphasized the reality of the obstruction created by the hypoplasia of the transverse arch. The ratio of transverse arch to ascending aorta has been used to assess the degree of transverse arch hypoplasia [13]. It is theorized that a reduction of more than 50% in the diameter of the transverse arch in comparison with the diameter of terminal end of the ascending aorta will probably create a real obstruction and may require surgical relief [15]. The caveat is that, because of the decreased aortic blood flow, the size of the ascending aorta itself may be smaller than normal in newborns presenting with coarctation. Comparison with the size of the descending aorta is more consistent because the diameter of the descending aorta is most often normal in neonates presenting with coarctation and reflects the body weight [10, 12]. Mee and colleagues of the Royal Children s Hospital in Victoria, Australia, described a unique formula. They considered the aortic arch to be hypoplastic if the transverse arch diameter is less than the patient s body weight in kilograms plus 1 [2]. The Z value, defined as the number of standard deviations from the mean normal expected size, can also be used to express the size of the hypoplastic aortic arch [7]. We consider the aortic arch to be hypoplastic if the Z value is 2 or more. Babies with such anatomy are considered unsuitable candidates for coarctation repair alone. We believe that the use of the Z value to express the aortic arch anatomy is more accurate and gives a clear impression of the size and growth potential of the arch. One-Stage Versus Staged Repair and Optimal Surgical Technique Neonatal coarctation of the aorta with arch hypoplasia seems to be unique among other congenital cardiac defects in that no other anomaly is managed with such diverse surgical techniques. All of the techniques used, with the exception of radically extended end-to-end anastomosis, can be done only through a left thoracotomy. This approach, of course, does not allow concomitant repair of associated intracardiac defects. Conte and associates [11], in reporting a large series in which they used extended end-to-end anastomosis, recommended complete repair during a one-stage procedure through a median sternotomy. Several other reports indicate that aortic arch obstruction with concomitant repair of associated cardiac defects is feasible and that it gives better results [2, 3, 9]. A high rate of recurrent coarctation has been reported after neonatal coarctation repair by end-to-end anastomosis [17], subclavian flap angioplasty [18], and synthetic patch aortoplasty [19]. Although it has been emphasized that extended end-to-end anastomosis relieves arch obstruction more effectively, the recoarctation rate is still in the range of 10% [11, 20]. In contrast, the recoarctation rate in our study after 5 years was only 2%. We believe there are several advantages to our approach. First, it allows one-stage complete repair of all lesions, in accordance with the recent trend in the literature. We perform complete resection of ductal tissue. It has been stressed that complete excision of the coarctation and bypass of

Ann Thorac Surg ELGAMAL ET AL 2002;73:1267 73 AORTIC ARCH ADVANCEMENT 1271 the isthmus, which frequently contains ductal tissue, are valuable techniques in avoiding secondary constriction of the aorta. Second, our technique avoids the use of prosthetic material or sacrifice of the arch vessels. Third, it allows complete and radical relief of the transverse arch obstruction, as we anastomose the descending aorta to the side of the terminal part of the ascending aorta. It has been emphasized that any successful technique should address complete relief of arch hypoplasia. The operative morbidity and mortality and recoarctation rate are higher if the hypoplastic segment is disregarded, and the resulting residual obstruction will compound left ventricular dysfunction [2]. Subsequent Growth of the Aortic Arch Whether the arch grows after repair of coarctation with aortic arch hypoplasia is highly debatable. Bruwer and associates [7] believe that the arch grows after simple resection and end-to-end anastomosis and that enlargement of the hypoplastic arch is not necessary. However, they also reported a recoarctation rate of 12.5%. Jahangiri and associates [8] believe that the subclavian flap repair is an effective technique and that the arch will grow in the majority of patients without additional procedures. On the other hand, Myers and associates [16] believe that whereas a smaller-than-normal aortic arch may grow after simple end-to-end repair and after subclavian and patch aortoplasties, a truly hypoplastic arch will not. Their study did not include infants with tubular hypoplasia of the aortic arch proximal to the left subclavian artery. Moreover, exercise testing detects hypertension and arch gradients in patients with a good coarctation repair as assessed at rest. The hypertension and arch obstruction appear to be related to discrepancies in the growth of the transverse aortic arch proximal to the repair site, rather than a recoarctation of the aorta [21]. Proirer and associates [22] reported a series of 37 patients who underwent surgical treatment of hypoplastic arches. Thirty of them (81%) had had previous interventions for repair of coarctation either by subclavian flap angioplasty or end-to-end anastomosis. This clearly indicates the failure of the hypoplastic arch to grow after repair. Histologic studies have confirmed that the hypoplastic arch has an abnormal structure. It is characterized by a significantly higher-than-normal ratio of elastin lamellae to vessel diameter and by increases in collagen and decreases in -actin positive cells that may hinder the ability of the arch to distend [23]. It is our belief that, to avoid residual coarctation and recoarctation, the initial repair technique should completely relieve any obstruction of the entire aortic arch. This concept is supported by the results of the present study, which included only one case of recoarctation (in a patient with Shone s complex). Study Limitations The present study has some limitations. These include the inevitable need for DHCA, with all of its potential disadvantages, including psychoneurologic complications; the short follow-up period; and the lack of a control group treated by alternative and less invasive techniques through left thoracotomy for comparing the rate of recurrence for different techniques. It is our intention, though, to follow-up the patients studied here, evaluate their psychoneurologic development and long-term recurrence rate, and report the findings. Conclusions We conclude that aortic arch advancement, as a surgical technique for the management of a special group of patients presenting with NcoAo/AAH, is associated with reasonable morbidity and mortality and a low recurrence rate. The procedure also allows concomitant repair of intracardiac defects. The late results appear favorable. However, longer follow-up is required to assess the long-term sequelae of DHCA and late recurrence of coarctation. Aortic arch advancement is the treatment of choice for NCoAo/AAH at our institution. References 1. Mustard WT, Rower RD, Keith JD, Sirek A. Coarctation of the aorta with special reference to the first year of life. Ann Surg 1955;141:249 52. 2. Karl TR, Sano S, Brawn W, Mee RBB. Repair of hypoplastic or interrupted aortic arch via sternotomy. J Thorac Cardiovasc Surg 1992;104:688 95. 3. Haas F, Goldberg CS, Ohye RG, Mosca RS, Bove EL. Primary repair of aortic arch obstruction with ventricular septal defect in preterm and low birth weight infants. Eur J Cardiothorac Surg 2000;17:643 47. 4. Hammon JW, Merrill WH, Prager RL, et al. Repair of interrupted aortic arch and associated malformations in infancy: indications for complete or partial repair. Ann Thorac Surg 1986;42:17 21. 5. Hart JC, Waldhausen JA. Reversed subclavian flap angioplasty for arch coarctation of the aorta. Ann Thorac Surg 1983;36:715 7. 6. Vincent JG, Daniels O, Van Oort A, Lacquet LK. Hypoplastic aortic arch with aortic coarctation: surgical correction. J Thorac Cardiovasc Surg 1985;89:465 8. 7. Brouwer MHJ, Cromme-Dijkhuis AH, Ebels T, Eijgelaar A. Growth of the hypoplastic aortic arch after simple coarctation resection and end-to-end anastomosis. J Thorac Cardiovasc Surg 1992;104:426 33. 8. Jahangiri M, Shinebourne EA, Zurakowski D, Rigby ML, Redington AN, Lincoln C. Subclavian flap angioplasty: does the arch look after itself? J Thorac Cardiovasc Surg 2000;120: 224 9. 9. Gaynor JW, Wernovsky G, Rychik J, Rome JJ, DeCampli WM, Spray TL. Outcome following single-stage repair of coarctation with ventricular septal defect. Eur J Cardiothorac Surg 2000;18:62 7. 10. Vouhe P, Trinquet F, Lecompte Y, et al. Aortic coarctation with hypoplastic aortic arch. J Thorac Cardiovasc Surg 1988; 96:557 63. 11. Conte S, Lacour-Gayet F, Serraf A, et al. Surgical management of neonatal coarctation. J Thorac Cardiovasc Surg 1995; 109:663 75. 12. Rudolph AM, Heymann MA, Spitznas U. Hemodynamic consideration in the development of narrowing of the aorta. Am J Cardiol 1972;30:514 25. 13. Moulaert AJ, Bruins CC, Oppenheimer-Dekker A. Anomalies of the aortic arch and ventricular septal defects. Circulation 1976;53:1011 5. 14. Bharati S, Lev M. The surgical anatomy of the heart in tubular hypoplasia of the transverse aorta (preductal coarctation). J Thorac Cardiovasc Surg 1986;91:79 85. 15. Lacour-Gayet F, Bruniaux J, Serraf A, et at. Hypoplastic

1272 ELGAMAL ET AL Ann Thorac Surg AORTIC ARCH ADVANCEMENT 2002;73:1267 73 transverse arch and coarctation in neonates. Surgical reconstruction of the aortic arch: a study of sixty-six patients. J Thorac Cardiovasc Surg 1990;100:808 16. 16. Myers JL, McConnell BA, Waldhausen JA. Coarctation of the aorta in infants: Does the aortic arch grow after repair? Ann Thorac Surg 1992;54:869 75. 17. Tawes RL, Aberdeen E, Waterston DJ, Carter REB. Coarctation of the aorta in infants and children: a review of 333 operative cases including 170 infants. Circulation 1969; 39(Suppl I):173 84. 18. Metzdorff MT, Cobanoglu A, Grunkemeier GL, Sunderland CO, Starr A. Influence of age at operation on late results with subclavian flap aortoplasty. J Thorac Cardiovasc Surg 1985; 89:235 41. 19. Hesslein PS, McNamara DG, Morris MJH, Hallman G, Cooley D. Comparison of resection versus patch aortoplasty for repair of coarctation in infants and children. J Thorac Cardiovasc Surg 1991;102:596 601. 20. Van Heurn LW, Wong CM, Spiegelhalter DJ, et al. Surgical treatment of aortic coarctation in infants younger than three months: 1985 to 1990;Success of extended end-to-end arch aortoplasty. J Thorac Cardiovasc Surg1994;107:74 85. 21. Weber HS, Cyran SE, Grzeszczak M, Myers JL, Gleason MM, Baylen BG. Discrepancies in aortic growth explain aortic arch gradients during exercise. J Am Coll Cardiol 1993;21: 1002 7. 22. Poirier NC, Van Arsdell GS, Brindle M, et al. Surgical treatment of aortic hypoplasia in infants and children with biventricular hearts. Ann Thorac Surg 1999;68:2293 7. 23. Machii M, Becker AE. Hypoplastic aortic arch morphology pertinent to growth after surgical correction of aortic coarctation. Ann Thorac Surg 1997;64:516 20. DISCUSSION DR KIRK R. KANTER (Atlanta, GA): On follow-up, did you notice any problems with the left pulmonary artery by bringing the descending aorta so proximally on the ascending aorta? DR ELGAMAL: No, we did not have any problem either left pulmonary artery or left main stem bronchus. DR CARL L. BACKER (Chicago, IL): I want to congratulate you on a very nice presentation. However, I would like to bring up a couple of important points. First, I disagree with the statement you made at the beginning of your talk that there is a high incidence of recoarctation using current surgical techniques. I recently compiled a collected series totaling 608 patients that have had coarctation repair with resection with extended end-to-end anastomosis from eight centers in Europe and the United States. The incidence of recoarctation was 8% in that group of patients: similar patients to your series including patients with normal intracardiac anatomy, patients with VSD, and patients with complex intracardiac anomalies. The second thing I would like to point out is there is some advantage to avoiding cardiopulmonary bypass and circulatory arrest for the patients that do not have any intracardiac anomalies. I think historically these patients can be very well repaired through a left thoracotomy approach, and that avoids a blood transfusion in most cases, and it certainly avoids the risks of circulatory arrest. The question I would ask: What do you think the risk to these patients is of undergoing 30 minutes of circulatory arrest compared with no circulatory arrest? The final two questions relate to our series of coarctation repairs with resection and extended end-to-end anastomosis. When we looked at operating on coarctation from a median sternotomy approach, there was a higher incidence of temporary recurrent laryngeal nerve injuries. Did you look at the incidence of vocal cord problems after your procedures? Finally, did you have any patients with left bronchial obstruction from the tension caused by bringing the descending aorta so far proximal on the ascending aorta? Again, congratulations on a very nice presentation. DR ELGAMAL: Thank you very much for your nice comments. But as I said before, we do not have any case of left main stem bronchus obstruction, and we did not have any patient with recurrent laryngeal nerve or phrenic nerve injury after surgery. I do respect your point: Is it justifiable to put the patient with isolated coarct and arch hypoplasia on pump and circulatory arrest? I think yes, because our mean circulatory arrest in group 1 was 17 minutes. The range was 14 to 21. And I think it may be very hard to put the aortic cross-clamp on the arch, you have to occlude at least 50% of the innominate artery in order to bring the descending aorta to the concavity of that thing, and I think you may have residual arch obstruction in the proximal arch. DR TOM R. KARL (San Francisco, CA): I enjoyed your talk. I have used this technique extensively myself and I like it a lot, but I have the same concern as Carl Backer about the routine use of this strategy for patients who could have a repair through a left thoracotomy. I have two questions: (1) Have you explored any ways to use the transsternal approach but avoid circulatory arrest? (2) Whether or not you are using this operation for all patients now or if you have some selection criteria to stratify into left thoracotomy approach versus transsternal approach? I did not get that from the abstract nor from the presentation. DR ELGAMAL: Actually, in that series, we did not use any way to avoid circulatory arrest. But that technique we use in the Norwood procedure, we put the shunt first and then perfuse the brain through the shunt and avoid total circulatory arrest. DR CHRISTO I. TCHERVENKOV: (Montreal, Quebec, Canada): I would like to make a couple of comments and ask a couple of questions. This is obviously a very important area of discussion. We have had an interest in the patients with complex congenital heart disease and aortic arch obstruction. Several years ago, we reported our experience at the Montreal Children s Hospital with single-stage intracardiac repair and concomitant pulmonary homograft patch aortoplasty of the aortic arch. I congratulate you and Dr Fraser on your outstanding results with your surgical technique and approach. However, I feel that we have to begin looking at specific subsets of this heterogeneous group of patients in order to decide the optimal surgical approach for each group. Echoing somewhat what Carl Backer and Tom Karl said earlier, I have a little trouble with the group 1 patients with isolated coarctation and tubular hypoplasia with your approach. Other authors have reported techniques of end-to-side anastomosis to the proximal aortic arch. So, it is not clear to me why you chose to go all the way to the ascending aorta for these patients. For the complex patients with aortic arch obstruction, we have experience with over 50 patients, two-thirds of which had transposition complexes or hypoplastic left heart complex with multiple left-sided obstructions and left ventricular hypoplasia, with a 3.8% mortality. We feel that the pulmonary homograph

Ann Thorac Surg ELGAMAL ET AL 2002;73:1267 73 AORTIC ARCH ADVANCEMENT 1273 patch aortoplasty allows a consistent and reliable enlargement of the aortic arch. And particularly for the patients undergoing a concomitant arterial switch operation, you avoid two circumferential anastomoses with excessive tension. I would like to know what was the break down within the different subgroups of intracardiac anomalies in the complex group of patients. How many patients, for example, had transposition complexes? Were all such patients in your institution treated with the single-stage approach and by the surgical technique of aortic arch advancement? What was the outcome in this subgroup of patients? I think the story is not as simple as it might seem at first glance. We really have to begin to stratify this complex group of patients. On the other hand, I fully endorse the single-stage approach, certainly for the group 3 patients with complex intracardiac disease and for the patient with VSD and coarctation. Although the Congenital Heart Surgeons Society study did not support this approach for the latter type of patient a few years ago, I think that that experience is probably outdated now. So can you give additional information on the diagnostic stratification of your cases? DR ELGAMAL: Regarding the first group, the Z value for the proximal arch in group 1 was minus 5. So that is why we were aggressive with that. And we do not have any mortality or morbidity with this particular group of patients. The mean circulatory arrest time was 17 minutes, as I said before. And we think this procedure is correct because we do not have any single case of recurrence in this group of patients. Regarding the complex congenital, we have 5 patients with transposition of great arteries, and we do not have any recurrence in this group of patients. DR TCHERVENKOV: Do you currently treat all patients with transposition complexes with arch obstruction with this singlestage approach? DR FRASER: Maybe I ll just answer a couple of those, if that is all right, so Adel does not have to suffer all the arrows. As I am sure everyone that treats coarctation would attest, there are patients with proximal arch hypoplasia that we cannot address through the left chest. And this was our desire with this technique, to try and avoid that. We have certainly been presented with patients late after what would have been considered a very successful-coarctation repair with proximal arch hypoplasia. And it is my opinion that through the left chest you can only reach so far on the proximal arch, and this is what Adel is alluding to. It is our strategy to repair all of these patients in one stage through the front. However, on the isolated coarctation patients, we are selective, based on the size of the proximal arch, the base of the proximal arch distal to the innominate, and base on that, we use a Z score. I know, Tom, you have used the patient s weight as a different criterion. But we feel that the tissue-totissue repair is ideal and hopefully will be borne out over time. But certainly I agree that the answer is not in yet on that.