Ann Thorac Surg 2010;90:1996-2000. doi:10.1016/j.athoracsur.2010.07.038
© 2010 The Society of Thoracic Surgeons
Original Articles: Pediatric Cardiac
Changes of Brain Magnetic Resonance Imaging Findings After Congenital Aortic Arch Anomaly Repair Using Regional Cerebral Perfusion in Neonates and Young Infants
Jae Gun Kwak, MDa,
Woong-Han Kim, MDb,*,
Jin Tae Kim, MDc,
In-One Kim, MDd,
Jong-Hee Chae, MDe
a Department of Thoracic and Cardiovascular Surgery, Sejong General Hospital, Bucheon, Korea
b Department of Thoracic and Cardiovascular Surgery, Seoul National University Children's Hospital, Seoul, Korea
c Department of Pediatric Anesthesiology and Pain Medicine, Seoul National University Children's Hospital, Seoul, Korea
d Department of Pediatric Radiology, Seoul National University Children's Hospital, Seoul, Korea
e Department of Pediatrics and Adolescent Medicine, Seoul National University Children's Hospital, Seoul, Korea
Accepted for publication July 14, 2010.
* Address correspondence to Dr Woong-Han Kim, Department of Thoracic and Cardiovascular Surgery, Seoul National University Children's Hospital, Yongeon dong, Jongro gu, Seoul, Republic of Korea (Email: woonghan{at}snu.ac.kr).
 |
Abstract
|
|---|
Background: The objective of this prospective study is to compare magnetic resonance imaging (MRI) findings before and after surgery for repairing congenital aortic arch anomalies using regional cerebral perfusion.
Methods: Neurologic examinations that included brain MRI, brain sonography, and electroencephalograms were performed before and after surgery for congenital aortic arch anomalies and the accompanying intracardiac anomalies using regional cerebral perfusion in 11 neonates and young infants.
Results: The median age at operation was 11 days (range, 5 to 46). The diagnoses included coarctation of the aorta with accompanying intracardiac anomalies (n = 10) and interruption of the aortic arch (n = 1). Aortic arch repair was performed using regional cerebral perfusion through the right innominate artery (regional perfusion time: 25.6 ± 6.0 minutes) without cardiac arrest. Two patients had new postoperative lesions on postoperative brain MRI, and these were acute focal subdural hemorrhage (n = 1) and acute focal infarction (n = 1). However, they were without clinical significance. Periventricular leukomalacia was not observed on brain MRI. There was no significant change between the preoperative and postoperative findings on brain sonography and electroencephalograms. All the patients showed normal neurologic growth for a mean follow-up duration of 175.3 days (range: 25 to 497 days).
Conclusions: There were newly developed lesions on the postoperative brain MRI in 2 of 11 patients, even though these patients showed normal brain sonography and electroencephalogram findings and normal neurologic development. Our regional cerebral perfusion protocol for aortic arch repair showed tolerable neurologic outcomes, and it did not induce periventricular leukomalacia.
|
Introduction
|
|---|
Even though there is debate over the benefits of regional cerebral perfusion when performing surgery for congenial aortic arch anomalies, such as coarctation of the aorta (CoA) or an interrupted aortic arch, many surgeons have applied their own unique regional cerebral perfusion technique for repairing congenital aortic arch anomalies. The objective of this study is to compare brain magnetic resonance imaging (MRI) findings before and after an operation for congenital aortic arch anomalies using our regional cerebral perfusion protocol, especially in neonates and young infants. There have been some studies to elucidate the advantages of regional cerebral perfusion for the neurologic outcomes, yet to the best of our knowledge, there have been few studies that have compared the brain MRI findings before and after cardiac surgery with using regional cerebral perfusion for congenital aortic arch anomalies [1–3]. We also tried to assess the advantages of our regional cerebral perfusion protocol for preventing the occurrence of periventricular leukomalacia (PVL) after the operation, which is known to occur at a high rate after cardiac surgery in neonates and young infants [1, 2].
|
Patients and Methods
|
|---|
From May 2008 to August 2009, 11 neonates (9 younger than 30 days) and young infants (n = 2, aged 38 days and 40 days) who underwent surgical repair of congenital aortic arch anomalies and their concomitant intracardiac anomalies were enrolled in this prospective study. The University Ethics Committee reviewed and approved this study (approval number H-0803-043-238), and fully informed written consent regarding the preoperative and postoperative neurologic study and evaluation, the intraoperative neurophysiologic monitoring, and the data collection with analysis was obtained from all the parents of the patients who participated in this study.
All the cardiovascular anomalies were totally repaired with one-stage surgery, except for 1 univentricular patient who underwent further palliation surgery 1 year after repairing CoA. All of the aortic arch anomalies were repaired using our regional cerebral perfusion protocol, with keeping the heart beating being allowed by coronary artery perfusion as well as by cerebral perfusion (dual regional perfusion). The detailed protocols were previously described [3, 4]. A 6F (2.0 mm) or 8F (2.7 mm) arterial cannula was inserted directly through the innominate artery, and standard bicaval cannulation was instituted. The aortic root cannula was connected to innominate arterial cannula with T-shaped connector (three-way connector) for dual perfusion to coronary artery to maintain heart beating during the repair of arch anomalies. We clamped and snared all arch vessels including the innominate artery, which was used for regional perfusion, and the ascending aorta was also clamped above the root cannula during the repair of arch anomalies. Flow rate was maintained at approximately 50 to 100 mL · kg–1 · min–1, mean blood pressure of the right radial artery was maintained 50 to 60 mm Hg. The pH-stat strategy was used exclusively for acid-base management, and core temperature was maintained at approximately 28°C. During the operation, we measured the regional cerebral oxygen saturation index in the right and left hemispheres using near-infrared spectroscopy (Invos; Somanetics, Troy, MI) in real time.
We tried to perform brain sonography, electroencephalography, and brain MRI in all the patients before and after the operation. The brain MRI was checked for all the patients preoperatively and postoperatively, and the images were interpreted by one pediatric radiologist (K.I.O.). A pediatric neurologist (C.J.H.) regularly checked the neurologic development status of all the patients during the follow-up period as well as during the immediate postoperative period in the hospital. The median overall follow-up duration was 5.0 months (range: 1 to 16 months).
|
Results
|
|---|
Patient Profile and Surgical Outcomes
The median age of the patients at the time of operation was 11 days (range: 5 to 46 days), the median body weight was 3.72 kg (range: 2.43 to 4.4 kg), and the body surface area was 0.23 m2 (range: 0.17 to 0.26 m2). The number of male patients was 10. The diagnoses included CoA with ventricular septal defect, atrial septal defect or atrioventricular septal defect (n = 7), CoA with supramitral ring (n = 1), CoA with Taussig-Bing anomaly (n = 1), interrupted aortic arch with ventricular septal defect (n = 1), and single ventricle accompanied with double-outlet right ventricle and CoA (n = 1, Table 1). Aortic arch repair was performed under regional cerebral perfusion through the right innominate artery with a beating heart and without cardiac arrest or total circulatory arrest in all the patients. The mean regional cerebral perfusion time was 25.6 ± 6.0 minutes. The mean blood flow through the right innominate artery was 52.5 ± 5.8 mL · kg–1 · min–1. The accompanying intracardiac anomalies were totally repaired in one stage with arch anomalies except for 1 single ventricle patient. This single ventricle patient presented with left ventricle type with double-outlet right ventricle and CoA. This patient underwent bidirectional cavopulmonary shunt and Damun-Kaye-Stansel procedure 1 year after undergoing CoA repair. The mean cardiopulmonary bypass (CPB) time was 173.3 ± 49.2 minutes, and the mean aortic cross-clamping time was 91.8 ± 124.6 minutes. There was no operative and late mortality during the follow-up period. There were no significant complications after operation except for chylothorax in 1 patient.
Neurologic Outcomes
The near-infrared spectroscopy showed consistent and even oxygen saturation index in the right and left hemispheres, and no abrupt change or desaturation in the oxygen saturation index during the operation.
Preoperatively, there was 1 patient who already showed abnormal findings on brain MRI: multifocal brain infarction and multifocal periventricular hemorrhage at both lobes and a small amount of subdural hematoma along the right occipital lobe. These findings did not change after the operation on brain MRI. Except for this patient, all the other patients showed normal findings on the preoperative brain MRI. Two patients showed newly developed lesions on the postoperative brain MRI. One patient (40 days old, CoA with Taussig-Bing anomaly; regional perfusion time 18 minutes, flow rate 47 mL · kg–1 · min–1, mean blood pressure 44 mm Hg, core temperature 27.8°C) showed acute infarction in the right middle frontal gyrus, which was thought to have occurred owing to microembolic material or microair bubble, and another two dotlike acute infarctions in the right temporal lobe and right cerebellar hemisphere, which were thought to have occurred within 3 days after operation owing to the same reason, microembolism (Fig 1). Another patient (10 days old, CoA with ventricular septal defect; regional perfusion time 20 minutes, flow rate 56 mL · kg–1 · min–1, mean blood pressure 44 mm Hg, core temperature 28.2°C) showed focal linear lesion along the right tentorium, and this may have been due to the subdural hemorrhage (Fig 2). This lesion was also thought to be related to CPB. The operative data during the regional perfusion, such as regional flow rate and time, mean blood pressure, and core temperature, were not significantly different from other patients' data that did not show newly developed lesion on their postoperative brain MRI image. These 2 patients showed no clinical neurologic symptoms or signs such as seizures or abnormal movement in the immediate postoperative period in hospital and had a smooth postoperative course. None of patients showed PVL on brain MRI postoperatively, which is known to occur at a high rate after cardiac operations, and especially in neonates [1].
View larger version (96K):
[in this window]
[in a new window]
|
Fig 1. Comparison of the brain magnetic resonance image (MRI) findings between the preoperative (preop) period (left) and postoperative (postop) period (right) in patient 1. Three newly developed lesions after operation are shown on brain MRI in this patient. A focal wedge-shaped acute infarction lesion (A) in the right middle frontal gyrus, and two dotlike acute infarction lesions in (B) the right temporal lobe and (C) the right cerebellar hemisphere are shown in the postoperative image (white arrows).
|
|
View larger version (84K):
[in this window]
[in a new window]
|
Fig 2. Comparison of the brain magnetic resonance image findings between (A) the preoperative period and (B) the postoperative period in patient 2. A newly developed lesion, indicated by the white arrow in the right tentorium area (right) may be due to subdural hemorrhage. Owing to slight difference in slicing angle, it does not show the exact same cross-sectional level of imaging.
|
|
On the brain sonography, all the patients, except for 2 patients in whom we did not perform the examination, showed no differences between the preoperative and postoperative findings, even the patients who showed newly developed lesions on brain MRI. On brain sonography, there was no evidence of brain edema, which is one of the main concerns associated with regional cerebral perfusion, after the operations. We also checked the electroencephalogram of all the patients, except 1; the electroencephalograms did not show differences between the preoperative and postoperative findings. This was also true for the patients who showed newly developed lesions on brain MRI. During the follow-up period, the pediatric neurologist (C.J.H.) regularly checked all the patients, and all of them caught up to the normal growth level for their age group; that was also the case for the 2 patients who showed immediate postoperative newly developed lesion on brain MRI.
|
Comment
|
|---|
There are various causes of unexpected adverse neurologic outcomes after cardiac surgery in neonates and young infants. There are nonmodifiable factors such as genetic syndromes, but there are also modifiable factors such as perioperative hypoxic or ischemic insults, and intraoperative situations, including deep hypothermic total circulatory arrest [5]. The advances in CPB techniques, the myocardial protection protocols, and the surgical techniques and instruments have overcome these modifiable factors. Among these, the regional cerebral perfusion technique has shown the possibility of avoiding many adverse effects from deep hypothermic total circulatory arrest during the repair of aortic arch anomalies and arch reconstruction. Of course, there continues to be debate about the benefit of regional cerebral perfusion as compared with that of traditional total circulatory arrest [6–8]. There have been few published studies that have compared the brain MRI findings both before and after open heart surgery, and especially before and after repair of congenital aortic arch anomalies, using regional cerebral perfusion.
The concerns about regional cerebral perfusion are the development of brain edema due to continuous blood flow from CPB machine and additional inflammatory reactions due to the relatively prolonged CPB time, as compared with the total circulatory arrest technique [2]; and PVL, which is the necrosis of the deep white matter adjacent to the lateral ventricles caused by injury to immature oligodendroglial cells, is known to frequently occur (more than 50% of the patients) after cardiac surgery in neonates [1, 2]. The main causes of PVL are the ischemia or inflammatory reaction associated with CPB after cardiac surgery, as well as hypoxia, hypoglycemia, and meningitis in some patients [9, 10]. Immature oligodendrocytes and their progenitors are known to be susceptible to proinflammatory cytokines and circulating inflammatory mediators that are usually amplified during the CPB support, as well as because of the hypoxic status [11–13].
At this point, we confront the dilemma to choose one of the options to prevent PVL during open cardiac surgery (1) total circulatory arrest for avoiding a prolonged CPB time or (2) regional cerebral perfusion for avoiding ischemic insult. In our protocol for regional cerebral perfusion, we have used relatively high flow to the brain, more than 50 mL · kg–1 · min–1, as compared to the protocols of other centers. This may be supposed to induce brain edema due to the direct high flow to the brain and the various inflammatory cytokines induced by continuous CPB flow. However, with our protocol, there were no findings associated with brain edema or inflammatory reaction on the brain imaging study, the brain MRI, and especially on brain sonography, which is well known to show the acute brain edematous lesions immediate after insults.
Licht and colleagues [14] found that the cerebral blood flow is reduced in patients with PVL, so, from that point of view, our relatively high blood flow may be beneficial to prevent the development of PVL rather than low blood flow. The only newly developed lesions were focal infarction and subdural hemorrhage, and these lesions did not show the any clinical correlations with patients' neurologic development during the follow-up period, including the immediate postoperative period. One thing to be considered for these lesions is the location of lesions. All these lesions occurred in the right side of the hemisphere, which is the same side of the source of regional cerebral perfusion flow from the right innominate artery. Even though there was no evidence, we assume the possibility of relationship between the location of these lesions and high flow blood flow from the right innominate artery.
Another discussion point for regional cerebral perfusion is whether this method supplies adequate blood flows to the left side of brain or even blood flows to the both sides, left and right, of the hemispheres, because only the right innominate artery has been mainly used for regional cerebral perfusion in most cases. A previous study by the authors has already observed the even distribution to the right and left hemisphere through the right innominate artery during regional cerebral perfusion [3]. The actual amount of brain perfusion by autoregulation from a single source that is shared with the coronary blood flow can be another discussion point, because the actual amount of blood flow cannot be measured. This uncertainty can be one of the reasons for our "high blood flow" protocol, because there is possibility of lower blood flow to the brain in unexpected situations.
Owing to the small number of patients, we were not able to confirm our results with statistical methods; and we have no control group for our patients, for example, no total circulatory arrest group. However, as far as we expect our regional cerebral perfusion protocol has more advantages comparing to total circulatory arrest, it can be an unethical attitude toward the patient to arrange the patients into two groups, regional cerebral perfusion group and total circulatory arrest group. Even though this study was mainly designed to evaluate the perioperative change of brain lesions as seen on brain MRI, and especially for the aspect of newly occurring PVL during the early postoperative period, further follow-up is needed so we can evaluate these patients with more meticulous neurologic testing and tools. We are expecting this will make our results clearer.
In conclusion, there were newly developed lesions seen on the postoperative brain MRI in 2 of the 11 patients, however, these patients showed normal brain sonography and electroencephalogram findings and normal neurologic development. The regional cerebral perfusion protocol we used for anomalous aortic arch repair showed tolerable neurologic outcomes on imaging study, and there was no PVL among the patients, as is known to frequently occur after cardiac surgery in neonates.
|
References
|
|---|
- Galli KK, Zimmerman RA, Jarvik GP, et al. Periventricular leukomalacia is common after neonatal cardiac surgery J Thorac Cardiovasc Surg 2004;127:692-704.[Abstract/Free Full Text]
- Gaynor JW. Periventricular leukomalacia following neonatal and infant cardiac surgery Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2004;7:133-140.[Medline]
- Kwak JG, Kim WH, Oh AY, et al. Is unilateral brain regional perfusion neurologically safe during congenital aortic arch surgery? Eur J Cardiothorac Surg 2007;32:751-755.[Abstract/Free Full Text]
- Lim HG, Kim WH, Jang WS, et al. One-stage total repair of aortic arch anomaly using regional perfusion Eur J Cardiothorac Surg 2007;31:242-248.[Abstract/Free Full Text]
- Hsia TY, Gruber PJ. Factors influencing neurologic outcome after neonatal cardiopulmonary bypass: what we can and cannot control Ann Thorac Surg 2006;81(Suppl):2381-2388.
- Ungerleider RM, Gaynor JW. The Boston Circulatory Arrest study: an analysis J Thorac Cardiovasc Surg 2004;127:1256-1261.[Free Full Text]
- Wypij D, Newburger JW, Rappaport LA, et al. The effect of duration of deep hypothermic circulatory arrest in infant heart surgery on late neurodevelopment: the Boston Circulatory Arrest Trial J Thorac Cardiovasc Surg 2003;126:1397-1403.[Abstract/Free Full Text]
- Bellinger DC, Wypij D, duPlessis AJ, et al. Neurodevelopmental status at eight years in children with dextro-transposition of the great arteries: the Boston Circulatory Arrest Trial J Thorac Cardiovasc Surg 2003;126:1385-1396.[Abstract/Free Full Text]
- Banker BQ, Larroche JC. Periventricular leukomalacia of infancy. A form of neonatal anoxic encephalopathy. Arch Neurol 1962;7:386-410.[Medline]
- Miller SP, Shevell MI, Patenaude Y, O'Gorman AM. Neuromotor spectrum of periventricular leukomalacia in children born at term Pediatr Neurol 2000;23:155-159.[Medline]
- Volpe JJ. Neurologic outcome of prematurity Arch Neurol 1998;55:297-300.[Medline]
- du Plessis AJ, Volpe JJ. Perinatal brain injury in the preterm and term newborn Curr Opin Neurol 2002;15:151-157.[Medline]
- Inder TE, Volpe JJ. Mechanisms of perinatal brain injury Semin Neonatol 2000;5:3-16.[Medline]
- Licht DJ, Wang J, Silvestre DW, et al. Preoperative cerebral blood flow is diminished in neonates with severe congenital heart defects J Thorac Cardiovasc Surg 2004;128:841-849.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
H. Dave, B. Rosser, K. Reineke, S. Nguyen-Minh, W. Knirsch, and R. Pretre
Aortic arch enlargement and coarctation repair through a left thoracotomy: significance of ductal perfusion
Eur J Cardiothorac Surg,
April 1, 2012;
41(4):
906 - 912.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
