|
|
||||||||
Eur J Cardiothorac Surg 2003;23:149-155
© 2003 Elsevier Science NL
Department of Cardiovascular Surgery, Sejong General Hospital, Sejong Heart Institute, Bucheon, Kyungki-do, South Korea
Received 24 June 2002; received in revised form 11 October 2002; accepted 21 October 2002.
* Corresponding author. Department of Cardiovascular Surgery, Sejong General Hospital, 91-121 Sosa Bon 2-dong, Sosa-ku, Bucheon-shi, Kyungki-do, 422-232, South Korea. Tel.: +82-32-340-1151; fax: +82-32-340-1236
e-mail: woonghan{at}korea.com
| Abstract |
|---|
|
|
|---|
Key Words: Aortic arch repair Regional perfusion
| 1. Introduction |
|---|
|
|
|---|
The purpose of our study was to describe our surgical technique using LFRP and to investigate its effectiveness in view of cerebral and myocardial protection.
| 2. Materials and methods |
|---|
|
|
|---|
|
When the rectal temperature reached 18 °C, the proximal innominate artery, left common carotid artery, and left subclavian artery were snared down to initiate the regional cerebral perfusion. Ascending aorta just distal to the root cannula was clamped for isolated myocardial perfusion. Descending aorta was also clamped as far distally as possible and gently elevated to produce a bloodless field and to reduce the anastomotic tension. The flow rate was regulated at about 50100 ml/kg per min. The mean blood pressure of the right radial artery was maintained at about 4050 mmHg and the mean hematocrit was maintained at around 20%. Simultaneous myocardial perfusion was maintained using a T-connected infusion line. Aortic arch repair was performed with extended end-to-side anastomosis of the descending aorta to the ascending aorta and aortic arch using a native tissue-to-tissue technique, which was described by Fraser and Mee [15] and Rajasinghe et al. [16] (Fig. 1) . In patients with HLHS, a triple anastomosis technique was used, which consisted of descending aorta, ascending aorta, aortic arch, and main pulmonary artery anastomosis [15].
|
2.3. Statistical analysis
Statistical analysis was performed with the Statistical Analysis System software package (version 6.12; SAS Institute, Cary, NC). The significance of differences between two groups was assessed by unpaired Student's t-test,
2-test or Fisher's exact test. All results were expressed as the mean±standard deviation and a value of P less than 0.05 was considered statistically significant.
| 3. Results |
|---|
|
|
|---|
|
Operative mortality rates in each group were 0 and 18.2% (0/26 and 4/22, P=0.02). The first patient was a 19-day-old female neonate weighing 3.2 kg who had unbalanced atrioventricular septal defect (AVSD) with severe arch hypoplasia. Due to the closure of PDA, metabolic acidosis and eventual cardiac arrest occurred preoperatively. She underwent emergent coarctoplasty, pulmonary arterial internal banding, and atrial septectomy. However, she suffered from acute renal failure, sepsis, and subsequent multiorgan failure postoperatively, and died. The second patient was a 12-day-old female neonate weighing 3.4 kg who had posterior transposition of the great arteries with severe arch hypoplasia. She also underwent palliative coarctoplasty with pulmonary arterial internal banding and atrial septectomy. Several days later, she underwent an arterial switch operation. She could not be weaned from CPB and we applied an extracorporeal membrane oxygenator (ECMO) for 16.3 h, but in vain. A coronary artery problem seemed to be the cause of death. The third patient was a 10-day-old female neonate weighing 2.7 kg who had HLHS with mitral atresia, VSD, and obstructive total anomalous pulmonary venous return (TAPVR). She underwent a modified Norwood procedure with TAPVR repair. Postoperatively, she showed low cardiac output syndrome and eventually died after 3 h, maybe due to the pulmonary hypertensive crisis. The last patient was an 11-day-old male neonate weighing 3.2 kg who had HLHS with mitral and aortic atresia. He underwent a modified Norwood procedure and died on the first operative day because of low cardiac output syndrome.
Follow-up of all (44/48) operative survivors was completed for the mean duration of 9.1±5.4 months (range 0.121 months). The mean duration of hospital stay was 28.8±22.1 days (range 12128 days). Late mortality rates were 0 and 11.1% (0/26 and 2/18) in each group (P=0.08). One patient was a 10-month-old male infant weighing 6.5 kg who had functional single ventricle, criss-cross heart, subaortic stenosis, and type B interruption of the aortic arch. He underwent palliative aortic arch repair with conal septum resection, atrial septectomy, and pulmonary arterial internal banding. About 6 months after the first operation, he underwent bidirectional cavopulmonary shunt (BCPS) and died from pulmonary arterial thromboembolism about 3 months after the second operation. The other patient was a 29-day-old male neonate weighing 2.9 kg who had unbalanced AVSD, parachute left sided AVV with moderate regurgitation, and severe arch hypoplasia. He underwent palliative coarctoplasty with pulmonary arterial internal banding. About 7 months after the first operation, he underwent BCPS and subsequently received AVV repair and replacement due to progressive AVV regurgitation on the 27th and the 56th postoperative days, respectively. Eighteen days after the fourth operation, he expired due to low cardiac output syndrome.
Complications consisted of low cardiac output in eight cases, wound problems including one delayed sternal closure in five cases, sepsis in three cases, transient neurological problems in two cases, and transient renal insufficiency in two cases, respectively (Table 2). Among eight patients who showed postoperative low cardiac output syndrome, two had ECMO support. One in group I survived and the other in group II expired. Neurological complications occurred in two patients (4.2%). One infant with hemitruncus (right pulmonary artery from aorta) in group I showed transient choreoathetosis-like movement but completely recovered after 11 months of follow-up. The other patient with hypoplastic left heart complex and preoperative sepsis showed seizure-like movement but improved after 3 months of follow-up. Renal insufficiency occurred in two patients in group I, but the patients recovered shortly after peritoneal dialysis.
Recurrent aortic arch obstruction occurred in one patient. He was an 8-day-old neonate weighing 3.1 kg who had HLHS with aortic atresia, diminutive ascending aorta (diameter 1.52 mm), retrograde ascending aortic flow, severe arch and isthmic hypoplasia, large VSD, normal mitral valve, and normal sized left ventricle. He underwent one-stage biventricular repair that consisted of a modified Norwood procedure and Rastelli operation using homemade non-valved conduit. Five months after the operation, recoarctation developed and the pressure gradient was measured at about 74 mmHg. Successful balloon dilatation was performed and the pressure gradient dropped to 23 mmHg.
| 4. Discussion |
|---|
|
|
|---|
Postoperative neurological complications after deep hypothermic TCA were well recognized and several authors proposed new techniques to avoid those potential side effects of TCA. Immediate postoperative and long-term neuro-developmental complications were systematically analyzed by the Boston Circulatory Arrest Group. However, only short-term results using the regional perfusion technique without circulatory arrest were reported. In our study, both early and late mortality rates were acceptable and significantly different in both simple and complex groups. Early postoperative neurological complications occurred in two patients (4.2%), which was also acceptable compared to the normally reported rates of 425% in TCA groups [17]. Fortunately, they recovered quickly during follow-up, but the long-term complications such as neuro-developmental abnormalities need to be evaluated further. In our opinion, the rate of neurologic complication was not so high. The number was only two, without significant statistical meaning. Individual characteristics such as associated anomalies in the central nervous system, poorly developed circle of Willis or other collaterals, or thromboembolic events seemed to be responsible for those neurological problems. Some reports showed no neurologic complication at all, but we could not fully understand those remarkable results. They could have occurred by chance. Our two patients had no abnormality on CT scanning of the brain and electroencephalogram. And they had fully recovered during the follow-up period, as described above. Actually we cannot imagine the cause of neurologic deficits, whether it be the improper technique (i.e. hyperperfusion) or any other factors such as preoperative hypoxic insult or poor collaterals.
Besides the use of regional perfusion, other factors such as pH-stat strategy for acid-base management, deep hypothermia, improved surgical skills, and comprehensive perioperative intensive care contributed to our improved short-term results. Recently, Jonas [3] emphasized that deep hypothermic circulatory arrest would be used more safely than innovative but unproven methods of regional perfusion unless clear guidelines regarding flow rate, temperature, and duration were established.
The optimal range of cerebral blood flow in deep hypothermic milieu (18 °C) is still debatable. Initially, Asou et al. [7] and McElhinney et al. [8] used flow rates of 50 and 30 ml/kg per min, respectively. Pigula et al. [18] objectively quantified that 20 ml/kg per min is sufficient using near-infrared spectroscopic analysis (NIRS). In a recent review, Pigula [19] has liberalized flow rates up to 40 ml/kg per min for both cerebral and somatic circulatory support. The neonates usually had larger brain volume than that of adults, weighing one-seventh to one-tenth of the body weight. Moreover, they had more extensive collateral circulation to the opposite side of the body including somatic circulation. Additionally, we used simultaneous myocardial perfusion combined with cerebral perfusion. It needed a higher flow rate than isolated cerebral perfusion (an additional myocardial blood flow of about 15 ml/kg per min; Sano and Mee [11]). Ishino et al. [13] used a flow rate of 3050% of full flow and maintained arterial pressure at 3045 mmHg using combined cerebral and myocardial perfusion in moderate hypothermia (28 °C). We used a somewhat higher flow rate of 50100 ml/kg per min, which was similar to that of Ishino's report [13]. It is well known that the cerebral blood flow is not related to rate, but related to pressure [20]. Therefore, the optimal flow rate during our technique might be much higher than that suggested above, in order to maintain perfusion pressure at about 4050 mmHg. We considered that both cerebral and myocardial perfusion were effectively maintained in this higher flow rate. The risk of cerebral edema could be reduced using careful monitoring of right radial artery pressure and modified ultrafiltration. In our opinion, the higher the flow rate is, the better the clinical outcomes will be in terms of recovery of vital organ function as long as cerebral hyperperfusion does not occur. There is no evidence of definite cerebral hyperperfusion syndrome in our study populations.
Pigula et al. [21] reported that the somatic perfusion through various collaterals was maintained during LFRP with a flow rate of 3040 ml/kg per min and suggested that the support of the subdiaphragmatic viscera should improve the ability of neonates to survive the postoperative period. Because of relatively rich collaterals and intact circle of Willis in the neonates, perfusion to the subdiaphragmatic viscera and left side of the brain was sufficient with such a low flow rate even if the mean perfusion pressure of the umbilical artery was maintained at 12 mmHg. In our study, the duration of regional perfusion was 33±16 min, which was shorter than their report. However, the mean perfusion pressure of the right radial artery was maintained at about 4050 mmHg, which was rather higher than theirs. We did not uniformly use the left radial artery pressure monitoring line. Usually there is some difficulty in peripheral arterial cannulation in small neonates or infants, and we cannot choose the monitoring site liberally. However, in our experience, an abundance of collaterals made it possible to maintain the perfusion pressure over 30 mmHg in the contralateral side. Renal insufficiency occurred in only two patients (4.2%) and, as described earlier, they recovered soon after the peritoneal dialysis. But the direct comparison of their results showing no renal complication was inappropriate because of the difference in the sample size and various preoperative conditions. We considered that such a high flow rate would be beneficial in the functional recovery of somatic organs as well as cerebral protection.
Recent studies suggested that the higher hematocrit level would be beneficial on neurologic outcomes after the TCA or low flow bypass [22]. But there was some debate on the optimal hematocrit level and there has been no gold standard until now. We still believed that the lower hematocrit facilitated the cerebral blood flow and oxygen delivery with its rheologic benefits.
In our study, the rectal temperature was maintained at about 18 °C. We agreed that the minimal temperature during regional cerebral perfusion would not be too profound as deep hypothermic circulatory arrest. Therefore, we used core-cooling as low as 28 °C, similar to the usual moderate hypothermic CPB. But we used a unique separate air-cooling system of the operating room in order to protect the myocardium during cardioplegic arrest. So we did not use any topical myocardial cooling such as ice-slushed cold saline in order to prevent any cold injury to the myocardium or phrenic nerves. As a result, a low room temperature (about 10 °C) inevitably reduced the core temperature of the small neonate or infant because of the surface cooling effect. Moreover, a lower temperature could protect the spinal cord during cross-clamping of the descending thoracic aorta. We believe that the deep hypothermia had beneficial effects on both myocardial and spinal protection, in addition to cerebral protection. We used both rectal and nasopharyngeal temperature only. Tympanic temperature cannot predict the brain temperature accurately [23]. It also has unacceptably low sensitivity, specificity, positive predictive value and negative predictive value in neonates and infants [24].
We exclusively used a direct cannulation method through the innominate artery even in the small neonates instead of an indirect method using synthetic graft made of polytetrafluoroethylene. We considered it a very simple and reproducible method which did not obscure the surgical field. In addition, there was no cannulation-related complication such as stenosis or intimal injury. Usually, the size of the innominate artery was sufficient for cannulation and we easily inserted the 6Fr or 8Fr arterial cannula (RMI®, Edwards Lifesciences LLC, Irvine, CA). Even in an 18-day-old full-term neonate weighing 1.6 kg who had a functional single ventricle with subaortic obstruction, we introduced a 16-gauze angiographic catheter into the innominate artery without any problem. He received a modified DamusKayeStansel procedure 6 months after the first operation and subsequently received BCPS 14 months later. He is now waiting for a Fontan operation and angiography shows a widely patent innominate artery. For neonates with HLHS undergoing stage I Norwood palliation, we moved the cannula to the neo-aorta and constructed the shunt at the more distal innominate artery. We believe that any injury to the inner surface of a small polytetrafluoroethylene graft can compromise the shunt function, especially thrombotic occlusion. Therefore, we did not cannulate the shunt directly.
We used separate myocardial perfusion during aortic arch reconstruction. In our deep hypothermic temperature setting, the heart maintained extremely slow and empty beating, showing no signs of dilatation or functional derangement. Actually we had no comparative data between myocardial perfusion and cardioplegic arrest. But we believed that our myocardial perfusion technique practically reduced the myocardial ischemic time and gave us more time during intracardiac repair.
The limitations of our study were the usual problems in retrospective studies, lack of precise neurological examination specific to the neonates or infants, and non-randomized uncontrolled patient groups. Because the time span using TCA is quite different from that of regional perfusion, it has little meaning in comparing the neurologic outcome between two techniques. To evaluate the long-term effectiveness of our technique on the neuro-developmental outcome, a prospectively randomized controlled study with precise neurological examination and perioperative monitoring is necessary.
In conclusion, regional perfusion is feasible and can be used with acceptable results, which are comparable to those reported from many institutions using circulatory arrest. It may reduce the potential complications following aortic arch reconstruction using circulatory arrest. However, repair of the aortic arch in patients with complex intracardiac defects still has a significant rate of mortality and morbidity.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Song, W.-H. Kim, H. Kim, J. Koh, J. G. Kwak, K. B. Kim, and E. J. Bae Repair of Coarctation in Right Circumflex Retroesophageal Arch Ann. Thorac. Surg., January 1, 2009; 87(1): 307 - 309. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Gargiulo, G. Oppido, E. Angeli, and C. P. Napoleone Neonatal aortic arch surgery MMCTS, July 23, 2007; 2007(0723): 2345. [Abstract] [Full Text] [PDF] |
||||
![]() |
H.-G. Lim, W.-H. Kim, W.-S. Jang, C. Lim, J. G. Kwak, C. Lee, S. W. Hwang, and C.-H. Lee One-stage total repair of aortic arch anomaly using regional perfusion Eur. J. Cardiothorac. Surg., February 1, 2007; 31(2): 242 - 248. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. E. Beygui, F. Esmailian, D. A. Rigberg, and H. Laks Repair of the symptomatic aberrant aortic arch aneurysm without hypothermic circulatory arrest Interactive CardioVascular and Thoracic Surgery, December 1, 2004; 3(4): 569 - 572. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. J. Myung, M. Petko, A. R. Judkins, G. Schears, R. F. Ittenbach, R. J. Waibel, and W. M. DeCampli Regional low-flow perfusion improves neurologic outcome compared with deep hypothermic circulatory arrest in neonatal piglets J. Thorac. Cardiovasc. Surg., April 1, 2004; 127(4): 1051 - 1057. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Ishino and S. Sano Aortic arch repair with a working beating heart in premature infants J. Thorac. Cardiovasc. Surg., November 1, 2003; 126(5): 1653 - 1654. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |