Single-stage repair of aortic coarctation with ventricular septal defect using isolated cerebral and myocardial perfusion q

Transcription

1 European Journal of Cardio-thoracic Surgery 17 (2000) 538±542 Single-stage repair of aortic coarctation with ventricular septal defect using isolated cerebral and myocardial perfusion q Kozo Ishino*, Masaaki Kawada, Hiroyuki Irie, Koichi Kino, Shunji Sano Department of Cardiovascular Surgery, Okayama University Medical School, Shikata-cho, Okayama , Japan Received 7 September 1999; received in revised form 11 January 2000; accepted 22 February 2000 Abstract Objective: To avoid hypothermic circulatory arrest, we have repaired aortic coarctation with ventricular septal defect (VSD) in a onestage procedure using an isolated cerebral and myocardial perfusion technique, and retrospectively compared this novel approach to the conventional two-stage approach. Methods: Between October 1991 and February 1999, 24 infants, aged 4±137 days (median, 27 days) and weighing 1.7±4.3 kg (median, 3.0 kg), underwent the repair of aortic coarctation with VSD either in one (group I, n ˆ 11) or two stages (group II, n ˆ 13). In Group I, an arterial cannula for cardiopulmonary bypass was inserted into the ascending aorta in six patients with coarctation only, or into a polytetra uoroethylene (PTFE) graft which was anastomosed to the innominate artery in the remaining ve who had hypoplastic arches. A cross-clamp was placed between the innominate and left carotid arteries. The bypass ow was reduced to 30±50% of full ow at 288C, thereby maintaining a radial artery pressure of 30±45 mmhg. At this point, the aortic coarctation was repaired by an endto-end arch anastomosis, while maintaining brain perfusion and with the heart still beating. In ve patients with hypoplastic aortic arches, the innominate artery proximal to the graft was then secured down and the arch anastomosis was extended to the distal ascending aorta, while providing isolated cerebral perfusion and cardioplegic arrest. After arch reconstruction was performed, the clamp was moved onto the ascending aorta, and the VSD was closed with systemic perfusion. In contrast, for group II patients, coarctation repairs were performed through a posterolateral approach, and existing VSDs were closed as secondary procedures. Results: The mean isolated cerebral and myocardial perfusion time for group I was 13 min (range, 7±20 min). The myocardial ischemic time did not differ between groups I and II (43 ^ 4 vs. 42 ^ 5 min, not signi cant). There were no hospital mortalities or neurological complications in either group, but one late death in each group. Conclusion: Single-stage repair of aortic coarctation with VSD does not increase myocardial ischemic time compared to the traditional two-stage approach. The isolated cerebral and myocardial perfusion technique may offer substantial brain and myocardial protection during aortic arch reconstruction. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Coarctation; Ventricular septal defect; Circulatory arrest; Cerebral perfusion 1. Introduction The optimal management of aortic coarctation with ventricular septal defect (VSD) remains controversial [1]. Since the introduction of the preoperative use of prostaglandin E 1 to maintain ductal patency [2], most infants now survive until they are hemodynamically stable enough to undergo surgical repair. The traditional two-stage operation includes the repair of coarctation, with or without pulmonary artery banding, through a left thoracotomy, and closure of the concomitant VSD as a secondary procedure. q Presented at the 13th Annual Meeting of the European Association for Cardio-thoracic Surgery, Glasgow, Scotland, UK, September 5±8, * Corresponding author. Tel.: ; fax: address: ishino@tb3.so-net.ne.jp (K. Ishino) Recently, encouraging results have been reported for a primary total correction procedure through a midline sternotomy [3,4]. With two different procedures now available, a debate exists concerning the relative bene t of the singlestage versus the two-stage repair for these lesions. Deep hypothermia and circulatory arrest have been used by many surgeons to support vital organs during aortic arch reconstruction. However, circulatory arrest prolongs myocardial ischemic time and, regardless of its duration, may be associated with transient cerebral dysfunction [5,6] and delayed psychomotor development [7]. In order to avoid circulatory arrest, we have employed an isolated cerebral and myocardial perfusion technique for the singlestage repair of aortic coarctation with VSD, and retrospectively compared this technique with the conventional twostage approach /00/$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S (00)

2 K. Ishino et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 538± Materials and methods 2.1. Patient population Between October 1991 and February 1999, 24 infants (13 boys, 11 girls), aged 4±137 days (median, 27 days) and weighing 1.7±4.3 kg (median, 3.0 kg), underwent the repair of aortic coarctation with VSD at Okayama University Hospital. The indication for operation in all cases was congestive heart failure. Eighteen patients (75%) received an infusion of prostaglandin E 1 to maintain ductal patency prior to surgery. Six patients (25%) necessitated inotropic support, and three (13%) required mechanical ventilation. Additionally, balloon aortoplasty was attempted in two patients before the de nitive operation. The 24 patients were divided into two groups, based upon the surgical approach employed. Eleven infants (group I) underwent a complete single-stage repair after Five of them had coarctation along with a hypoplastic arch [3]. The location of the VSD was perimembranous in seven patients, subpulmonary in three, and muscular in one. Associated anomalies included a critical aortic stenosis in one infant, Ebstein's anomaly in another, and atrial septal defects (ASD) in four patients. Thirteen infants (group II) underwent a two-stage repair before Hypoplastic arches were present in four patients. The VSD was perimembranous in seven patients, subpulmonary in four, and muscular in two. In addition, two patients had a subaortic stenosis, three had ASDs, and one had Ebstein's anomaly Operative technique Single-stage repair Arterial blood pressure monitoring lines were placed in the right radial artery and femoral artery in each patient preoperatively. Through a midline sternotomy, the thymus gland was excised. The aortic arch, its branches and the duct were dissected out. The left carotid and left subclavian arteries were looped with tourniquets, to allow blood ow control. An arterial cannula was inserted into the ascending aorta in patients with aortic coarctation only, or into a 3.5- mm polytetra uoroethylene (PTFE) tube (Gore-Tex, WL Gore & Associates, Inc, Flagstaff, Ariz.) which was anastomosed to the innominate artery in patients with coarctation plus a hypoplastic arch (Fig. 1B). Cardiopulmonary bypass was commenced by bicaval cannulation with a ow rate of 150 ml/min per kg. During the cooling phase, the descending thoracic aorta was extensively mobilized by blunt dissection as far distally as possible without division of the intercostal arteries. Once a rectal temperature of 288C was obtained, the left carotid artery and left subclavian artery were snared. Isolated cerebral and myocardial perfusion were established by clamping the aortic arch between the innominate artery and left carotid artery with a curved clamp (Fig. 2A). To Fig. 1. An arterial cannula is inserted: (A), into the ascending aorta for repair of aortic coarctation; or (B), into a 3.5-mm PTFE tube which is anastomosed to the innominate artery for repair of coarctation plus a hypoplastic arch. keep a radial artery pressure of 30±45 mmhg, the bypass was maintained at 30±50% of full ow during isolated cerebral and myocardial perfusion. The duct was ligated and an additional clamp was placed on the descending aorta. With brain perfusion and while the heart was still beating, the coarctation was widely resected with all ductal tissue. In the six patients who only had coarctations, a longitudinal incision was made in the undersurface of the arch, which was then anastomosed to the obliquely trimmed distal aorta in an end-to-end fashion with continuous 7±0 polypropylene suture. After completion of the arch repair, the descending aortic clamp and head vessel snares were removed. The arch clamp was then repositioned onto the ascending aorta to allow systemic perfusion, and a VSD was closed with cardioplegic arrest (Fig. 2B). In the ve patients with coarctation plus a hypoplastic arch, about two-thirds of the arch anastomosis was accomplished with isolated cerebral and myocardial perfusion (Fig. 3A). The innominate artery just proximal to the PTFE perfusion tube was snared to maintain brain perfusion. After cardioplegic arrest, the arch was Fig. 2. Isolated cerebral and myocardial perfusion are established by clamping the aortic arch between the innominate artery and left carotid artery. (A) Coarctation repair is carried out with brain perfusion and the heart beating. (B) The clamp is repositioned onto the ascending aorta, and the VSD is closed with cardioplegic arrest.

3 540 K. Ishino et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 538± Results 3.1. Operative outcome Fig. 3. (A) In a case of coarctation plus hypoplastic arch, about two-thirds of the arch anastomosis is accomplished with isolated cerebral and myocardial perfusion. (B) The innominate artery just proximal to the PTFE tube is snared, the arch is unclamped, and the arch anastomosis is extended with isolated cerebral perfusion. unclamped, the aortic incision was extended beyond the origin of the innominate artery, and the anastomosis was completed with isolated cerebral perfusion (Fig. 3B). The distal clamp and all snares were removed, bypass ow was returned to normal, and VSD closure followed. In this group, the VSD was closed with a patch in all but one patient who underwent a direct suture closure. Additional procedures included an aortic valvotomy, repair of the tricuspid valve, and patch closure of ASD in three separate patients, respectively Two-stage repair During the initial operation, the coarctation was repaired through a left lateral thoracotomy by subclavian ap aortoplasty in 11 patients, and by resection of the coarcted segment with an end-to-end anastomosis of the aorta in the other two patients. Associated pulmonary artery banding was performed in four patients. After a mean interval of 10 months (range, 12 days±25 months), the VSD was closed with a Dacron patch in all 13 patients, using standard cardiopulmonary bypass and cardioplegic arrest. Concomitant procedures were performed as indicated and included a patch closure of an ASD in one child, and repair of the tricuspid valve and resection of the subaortic stenosis in a child with Ebstein's anomaly Statistical methods The medical records, operative reports and echocardiographic results were reviewed. Follow-up was completed for all patients. The Mann±Whitney rank sum test was used for the comparison of continuous variables between the two groups. The level of statistical signi cance was set at P, 0:05. The designation P ˆ NS indicates `not signi cant'. The data are presented as means ^ standard deviations. The isolated cerebral and myocardial perfusion time in group I was 13 ^ 4 min (range, 7±20 min), and the additional isolated cerebral perfusion time required in the ve patients with hypoplastic aortic arches was 11 ^ 2 min (range, 10±13 min). Thus, the total cross-clamping time of the descending aorta was 17 ^ 4 min (range, 10±23 min). The myocardial ischemic time in group I did not differ from the time seen during the second operation in group II (43 ^ 4 vs. 42 ^ 5 min, P ˆ NS). There were no hospital mortalities or neurological complications in either group Follow-up The follow-up periods in groups I and II were 20 ^ 12 (range, 2±44 months) and 67 ^ 25 months (range, 13±97 months), respectively. There were two late deaths; one group I infant died from infective endocarditis 2 months after surgery and one group II patient died from dilated cardiomyopathy 4 months after VSD closure. Recurrent coarctation, de ned as a peak gradient exceeding 20 mmhg across the arch, has not developed in any group I patient, but one group II patient required a balloon angioplasty due to an arch gradient of 33 mmhg. Re-operation following repair of coarctation and VSD was performed in two group II patients. Speci cally, one patient required repair of the mitral valve, while the other necessitated resection of subaortic stenosis. 4. Discussion Hypothermic circulatory arrest is a commonly used perfusion strategy during aortic arch reconstruction. Although deep hypothermia provides signi cant protection from brain damage, substantial evidence has accumulated that neurologic injury can occur from circulatory arrest, and can lead to both early and late sequelae [6,7]. Another disadvantage of circulatory arrest is the prolongation of myocardial ischemia. This correlation between circulatory arrest time and myocardial ischemia was borne out by Karl et al. [3], who achieved a better outcome in infants undergoing aortic arch repair with circulatory arrest, in whom myocardial ischemic time was reduced by employing isolated myocardial perfusion [8]. Previous reports [9,10] have recommended one-stage repairs of aortic coarctation and VSD without circulatory arrest, using a left thoracotomy followed by a midline sternotomy performed during the same anesthesia period. By using the isolated cerebral and myocardial perfusion technique, aortic coarctation and VSD could be repaired more easily during a single operation through a single incision, without interruption of the brain perfusion and without increasing the myocardial ischemic time.

4 K. Ishino et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 538± The isolated cerebral and myocardial perfusion technique was rst reported by Asou et al. in 1996 [11]. Their group perfused the innominate artery either through a PTFE shunt for Norwood operation or by a direct cannulation using a specially designed thin-walled cannula. In the present series, the ascending aorta, when large enough, was cannulated in patients with coarctation only, while a long PTFE tube on the innominate artery was the arterial perfusion route in patients with coarctation plus a hypoplastic arch. Because of the exibility of the tube, the extended aortic arch anastomosis was carried out without interference from the cannula and without any undue tension around the anastomotic site. This arterial perfusion route is currently our routine technique for complex aortic arch reconstructions, and has been applied in two babies with hypoplastic left heart syndrome who weighed less than 2 kg; both infants survived. The appropriate perfusion rate for the brain and heart under hypothermic conditions remains unknown. Asou et al. [11] reported their experiences with isolated cerebral perfusion in neonates, in which they selected a ow rate of 50 ml/kg per min at 228C, with a resultant right radial artery pressure of 20 mmhg. Pigula et al. [12] employed the same perfusion technique except with ow rates of 15±20 ml/kg per min at 188C, and documented suf cient brain perfusion based on cerebral oxygen saturation and relative cerebral blood volume. In terms of investigating adequate cardiac blood supply, Sano and Mee [8] maintained perfusion pressures at 30±45 mmhg for isolated myocardial perfusion through the ascending aorta. In our study, the innominate artery (at 288C) supplied both cerebral and myocardial perfusion. Given the uncertainty of ow distribution to these two vital organs, we adjusted the ow rates to keep perfusion pressures at 30±45 mmhg during isolated cerebral and myocardial perfusion. The management of a hypoplastic aortic arch is a surgical challenge. Such a lesion has been found in 60±70% of infants undergoing coarctation repair [13,14]. Although the extended arch anastomosis may be feasible through the left thoracotomy in the majority of patients [13,14], it is sometimes dif cult to place the upper aortic clamp proximal to the hypoplastic segment without seriously compromising ow to the innominate artery, at least with this exposure. However, the midline sternotomy, used for the single-stage repair in this series, obviously provides better exposure of the proximal arch. In the ve patients with coarctation plus a hypoplastic arch, a large anastomosis extending to the distal ascending aorta was possible without the proximal aortic clamp, but with brain perfusion through the innominate artery. None of these ve patients have developed a recoarctation in the medium-term follow-up. Thus, the repair of a hypoplastic arch by sternotomy may compare favorably in this respect with the same repair performed via thoracotomy. Isolated cerebral and myocardial perfusion may offer signi cant brain and myocardial protection during aortic arch reconstruction, although we have not provided any direct evidence of organ perfusion. This perfusion technique has now been standardized and the results are reproducible for a variety of surgical lesions, further verifying the applicability of the procedure. In particular, we have performed many more complex aortic arch repairs than reported here using this perfusion method, mainly for infants with interruption of the aortic arch and VSD, Taussig±Bing type double-outlet right ventricle, or hypoplastic left heart syndrome. References [1] Congenital Heart Surgeons Society, Quaegebeur JM, Jonas RA, Weinberg AD, Blackstone EH, Kirklin JW. Outcomes in seriously ill neonates with coarctation of the aorta: a multiinstitutional study. J Thorac Cardiovasc Surg 1994;108:841±854. [2] Olley PM, Coceani F, Bodach E. E-type prostaglandins: a new emergency therapy for certain cyanotic congenital heart malformations. Circulation 1976;53:728±731. [3] Karl TR, Sano S, Brawn WJ, Mee RBB. Repair of hypoplastic or interrupted aortic arch via sternotomy. J Thorac Cardiovasc Surg 1992;104:688±695. [4] Conte S, Lacour-Gayet F, Serraf A, Sousa-Uva M, Bruniaux J, Touchot A, Planche C. Surgical management of neonatal coarctation. J Thorac Cardiovasc Surg 1995;109:663±675. [5] Greeley WJ, Kern FH, Ungerleider RM, Boyd III JL, Quill T, Smith LR, Baldwin B, Reves JG. The effect of hypothermic cardiopulmonary bypass and total circulatory arrest on cerebral metabolism in neonates, infants, and children. J Thorac Cardiovasc Surg 1991;101:783±794. [6] Newburger JW, Jonas RA, Wernovsky G, Wypij D, Hickey PR, Kuban KCK, Farrell DM, Holmes GL, Helmers SL, Constantinou JE, Carrazana EJ, Barlow JK, Walsh AZ, Lucius KC, Share JC, Wessel DL, Hanley FL, Mayer JE, Castaneda AR, Ware JH. A comparison of the perioperative neurologic effects of hypothermic circulatory arrest versus low- ow cardiopulmonary bypass in infant heart surgery. N Engl J Med 1993;329:1057±1064. [7] Bellinger DC, Jonas RA, Rappaport LA, Wypij D, Wernovsky G, Kuban KCK, Barnes PD, Holmes GL, Hickey PR, Strand RD, Walsh AZ, Helmers SL, Constantinou JE, Carrazana EJ, Mayer JE, Hanley FL, Castaneda AR, Ware JH, Newburger JW. Developmental and neurologic status of children after heart surgery with hypothermic circulatory arrest or low- ow cardiopulmonary bypass. N Engl J Med 1995;332:549±555. [8] Sano S, Mee RBB. Isolated myocardial perfusion during arch repair. Ann Thorac Surg 1990;49:970±972. [9] Doty DB, Lauer RM, Ehrenhaft JL. Congenital cardiac anomalies: one-stage repair in infancy. Ann Thorac Surg 1975;20:316±325. [10] Tiraboschi R, Al eri O, Carpentier A, Parenzan L. One stage correction of coarctation of the aorta associated with intracardiac defects in infancy. J Cardiovasc Surg (Torino) 1978;19:11±16. [11] Asou T, Kado H, Imoto Y, Shiokawa Y, Tominaga R, Kawachi Y, Yasui H. Selective cerebral perfusion technique during aortic arch repair in neonates. Ann Thorac Surg 1996;61:1546±1548. [12] Pigula FA, Siewers RD, Nemoto EM. Regional perfusion of the brain during neonatal aortic arch reconstruction. J Thorac Cardiovasc Surg 1999;117:1023±1024. [13] Vouhe PR, Trinquet F, Lecompte Y, Vernant F, Roux PM, Touati G, Pome G, Leca F, Neveux JY. Aortic coarctation with hypoplastic aortic arch: results of extended end-to-end aortic arch anastomosis. J Thorac Cardiovasc Surg 1988;96:557±563. [14] Lacour-Gayet F, Bruniaux J, Serraf A, Chambran P, Blaysat G, Losay J, Petit J, Kachaner J, Planche C. Hypoplastic transverse arch and

5 542 K. Ishino et al. / European Journal of Cardio-thoracic Surgery 17 (2000) 538±542 coarctation in neonates: surgical reconstruction of the aortic arch. J Thorac Cardiovasc Surg 1990;100:808±816. Appendix A. Conference discussion Dr A. Corno (Lausanne, Switzerland): Since you have a report from the North American Congenital Heart Surgeons Society showing that there is a signi cantly better survival at 2 years after a staged approach versus primary repair, based on what kind of hypothesis do you start your study with a primary approach? Are your series of consecutive patients or, during the same period, did you offer a staged approach to some patients? Dr Sano: The results of our series were also the same in a single-stage repair group and a two-stage repair group. Therefore, I think it depends on the surgeon's policy. We used a two-stage repair until 1995, and have changed to a single-stage repair because we have developed a cerebral perfusion technique. If you adopt a single-stage repair, then you need to worry about the brain ischemia and brain damage, because of circulatory arrest. Although the arrest time is relatively short at coarctation repair, I think there may have some brain damage even in the subclinical level. As we have developed a cerebral perfusion technique, we don't need to worry about brain ischemia. This is why we have changed to a single-stage repair. There is also no difference in myocardial ischemic time between the two groups. Dr D. Anderson (London, UK): I noticed that one of your techniques of cerebral perfusion is to apply a tube graft to the innominate artery for perfusion. Did you nd that there were any problems related to that, either at the completion of the operation, or in the perfusion through the innominate artery or the right arm as a consequence of that technique? Dr Sano: We adopted this technique in 1995 and used this technique more than 50 cases, including the Norwood procedure, without any complication so far. We put a shunt before going on bypass and measure a right radial artery pressure and a femoral artery pressure. Unless the patient has very severe hypoplasia of the aortic arch, the pressures in the two sites are almost always equal. Dr H. Uemura (Osaka, Japan): Recently, I prefer to maintain perfusion, not only to the brain, but also to the organs of the lower body. When establishing cardiopulmonary bypass, two arterial cannulae are inserted, one to the ascending aorta and the other directly to the descending aorta. With this option, the postoperative urine output and the ndings in liver enzymes were much better compared with those in the group of patients undergoing ischemia of the lower body. In this respect, I think that maintenance of perfusion to the lower body should be less invasive as well. What do you think about this option? Dr Sano: I think that is a very good idea. Although, in a coarctation or interruption group, the mean circulatory arrest time to the lower body is only 10±20 min. Therefore, there is not much difference in postoperative urine, liver function and kidney function. Certainly, in the Norwood procedure, the circulatory arrest to a lower body takes more than 30 min, even in a good hand. In these cases, perfusion to a lower body may help more. I agree with you.