Abstract
Cerebral ischemia or infarction caused by several equivocal mechanisms is a major complication after aortic arch replacement. Here, we report a 28-year-old male who underwent total replacement of the aortic arch and concomitant tributaries for hypoplasia of the transverse aortic arch and aortic branches. Continuous cerebral oxygen saturation (rSO2) monitoring was applied throughout the whole surgical course. According to the trend of rSO2, we could not only optimize the cerebral perfusion, but also confirm the patency of graft anastomosis. Therefore, monitoring rSO2 is very useful for determining cerebral perfusion during major surgery, especially in complicated repair of an aortic aneurysm, or replacement of the aortic arch and/or arch vessels.
Keywords
aorta, thoracic; aortic arch; cerebral ischemia; monitoring, intraoperative; postoperative complications;
1. Introduction
Injury to the central nervous system caused by hypoperfusion or embolism is one of the major causes of morbidity and mortality after aortic arch surgery.1 Cerebral injury also accounts for prolonged hospitalization and has a significant impact on the cost of treatment.2 Therefore, several methods, including adequate cerebral perfusion and profound deep hypothermic circulatory arrest are used to preserve brain function during cardiac and aortic surgery. Cerebral oximetry, which makes use of the near-infrared spectrum (NIRS), is a new, continual and noninvasive monitoring technique for determination of the balance of brain oxygen between supply and demand, and can be used to safeguard against cerebral ischemia during complicated cardiac surgery.3 Here, we report our experience of using cerebral oximetry to monitor intraoperative brain oxygen balance and to confirm the patency of bypass grafts in a patient undergoing total replacement of the aortic arch and concomitant tributaries because of hypoplasia of the transverse arch and attached vessels.
2. Case Report
A 28-year-old male was scheduled for total aortic arch replacement with concomitant replacement of tributary vessels because of a complicated aortic abnormality. He had developed progressive dyspnea and chest discomfort without any neurological symptoms over the previous year. Because of a large difference in arterial pressure between both arms, left subclavian artery stenosis or coarctation of aorta was suspected. He thus underwent magnetic resonance imaging which revealed hypoplasia of the transverse aortic arch, left common carotid artery (LCCA) and left subclavian artery (LSA) (Figure 1). He was admitted for surgical intervention under the diagnosis of hypoplasia of the transverse aortic arch and concomitant vessels. On the day before surgery, we thoroughly discussed the possible procedures for surgery with the cardiovascular surgeon in charge and informed the patient and his family of these during the pre-anesthesia visit.
After the patient arrived in the operation room, basic monitors such as an electrocardiograph, a noninvasive blood pressure monitor and a pulse oximeter were set for use. An arterial catheter was inserted into the right radial artery and right dorsalis pedis artery for continuous blood pressure monitoring and blood sampling, respectively. Before anesthesia induction, a noninvasive INVOS® 5100B Cerebral Oximeter (Somanetics Corp., Troy, MI, USA) was prepared for continuously monitoring cerebral perfusion. After disinfection, two self-adhesive disposable sensors (SomaSensor®; Somanetics Corp.) were attached to the front, one on the left and the other on the right. The baseline values of regional cerebral oxygen saturation (rSO2) on the left and right sides read 74% (LrSO2) and 85% (RrSO2), respectively. The critical threshold of oxygen saturation for the cerebral hemisphere was specified as 59% for LrSO2 and 68% for RrSO2 by a reduction of 20% from baseline.4
General anesthesia was induced with midazolam (3 mg), fentanyl (200 μg), propofol (140 mg), and rocuronium (50 mg). Tracheal intubation was performed without any difficulty. After anesthesia induction, LrSO2 and RrSO2 increased to 77% and 87%, respectively. General anesthesia was maintained with isoflurane in an air/O2 mixture. The concentration of end-tidal isoflurane was kept around 1.0−1.2 minimum alveolar concentration. A pulmonary arterial catheter was inserted and transesophageal echocardiography was carried out for continuous surveillance of cardiac function. Both the left and right rSO2 remained unchanged after induction.
The operation started with left femoral arterial cannulation for general retrograde perfusion, followed by right subclavian artery (RSA) cannulation for antegrade cerebral perfusion (Figure 2). After sternotomy and dissection of the surrounding tissues, the aorta and its branches were hooked. Then the ascending aortic root was punctured and a perfusion needle inserted. After cannulation of the right atrium for venous return, cardiopulmonary bypass (CPB) was established.
Download full-size image
Download full-size image
During the initiation of CPB, both the left and right hemispheric rSO2 decreased but were above the critical threshold. General anesthesia during CPB was maintained with 2% isoflurane through a heart-lung machine. The surgeon decided to carry out the surgery without arrest of the heart by cardioplegia but with the heart beating and with antegrade coronary perfusion through the aortic perfusion needle. The body temperature was then decreased and maintained at 32ºC. The LrSO2 and RrSO2 were finally maintained at 69% and 75%, respectively, a 5% and 10% reduction from baseline levels, at which point the pump flow was about 3.8 L per minute, and the mean arterial pressure was about 55 mmHg. Then, the distal descending aorta, proximal aortic root and innominate artery (IA) were clamped, and antegrade cerebral perfusion was achieved through RSA cannulation with perfusion flow of 0.8 L per minute. At that point, LrSO2 decreased to 61% (13% reduction from baseline), whereas RrSO2 decreased to 70% (15% reduction).
Thereafter, the section of the hypoplastic aortic arch (Figure 3A) was resected from the proximal aorta and the descending aorta below the coarctation (Figure 3B) and the vessels leading from the aortic arch were also clamped and resected (Figure 3C). A Hemashield four-branch graft was used to replace the diseased aortic arch and the branch grafts were anastomosed with the respective severed arch vessels (Figure 3D). The ascending aorta was first anastomosed with the proximal end of the main graft, and then the distal end of the main graft was anastomosed with the descending aorta. The anastomoses of the LSA, LCCA and IA were performed smoothly with the corresponding branches of the main graft. After anastomoses of the LSA and LCCA, the grafts were unclamped, and the LrSO2 and RrSO2 increased from 61% to 70% and from 69% to 74%, respectively. Antegrade cerebral perfusion from RSA cannulation was stopped only after anastomosis of the IA with the graft and declamping of the distal IA. The LrSO2 and RrSO2 further increased to 75% and 80%, respectively. Following deaeration and rewarming, the patient was weaned from CPB successfully without any requirement for inotropic agents, and the LrSO2 and RrSO2 read 82% and 85%, respectively. At the end of surgery, bilateral rSO2 values were considerably decreased. The LrSO2 and RrSO2 were 73% and 74%, respectively, at the time the patient was transferred to the ICU. We did not try to increase the rSO2 because the overall rSO2 was above the critical threshold, which signified sufficient cerebral perfusion during the successful surgery. The patient fully awakened 20 minutes following arrival in the ICU and was extubated soon after. He was discharged uneventfully on the 10th day of hospitalization. There were no postoperative neurological abnormalities such as sensory or motor disturbance, headache or infarction during a 6-month follow-up. The changes in bilateral rSO2 during the course of operation are summarized in Figure 4.
Download full-size image
Download full-size image
3. Discussion
Cerebral damage is a catastrophic complication after surgical repair of the aortic arch. Neurologic injury may result from prolonged ischemia, emboli, or severe malperfusion.5 In the past, several techniques had been developed to avoid embolism by atherosclerotic debris and to preserve cerebral function during aortic arch reconstruction, such as by deep hypothermic circulatory arrest with or without antegrade selective cerebral perfusion.6
Since the last decade, technological research has allowed the clinical application of NIRS to continuously and noninvasively monitor cerebral oxygen saturation through the scalp and skull, and thus provide accurate information about the balance between brain oxygen supply and demand.7 Intraoperative cerebral oxygen desaturation has been significantly related to postoperative cognitive dysfunction as well as prolonged hospitalization and ICU stay.8 A recent interventional study showed a significant reduction in the incidence of stroke by optimization of the oxygen supply/demand ratio to maintain baseline rSO2 values during CPB.9
In our institution, the use of cerebral oximetry monitoring has been used to assess the balance of cerebral oxygen supply and demand during CPB. One of the major concerns of practitioners with using cerebral oximetry is the lack of a simple and uniform threshold to demarcate pathological cerebral saturation. Due to a wide variation in baseline rSO2, 10,11 and as some patients may have low values of baseline rSO2 of less than 50 points, we used a percentage decrease from baseline rSO2 of 15−20% as the critical threshold indicating the occurrence of cerebral ischemia.4,12 An unusual change in rSO2 value at several crucial moments may be an early warning of possible catastrophic cerebral ischemia resulting either from malpositioning of the aortic cannula, inadequate cerebral perfusion flow via graft anastomoses, embolic events, or unilateral venous obstruction. In this case, the rSO2 of each side was above the critical threshold during the operation, and there was no suspected cerebral ischemic event. To the best of our knowledge, this is the first case report concerning serial changes in rSO2 during this kind of complicated surgery.
In clinical application, some degree of asymmetry (2−4 points) between the left and right rSO2 values can frequently be observed. However, larger baseline differences in rSO2 value between left and right may be caused by carotid or intracranial arterial stenosis,13 an old infarction, extracranial lesions,14 outside interference with an infraredemitting device15 or incorrect head positioning. In our case, the relatively larger baseline difference in rSO2 value between left and right (LrSO2/RrSO2: 74/85) could have been caused by the hypoplasia of the LCCA.
After induction of anesthesia, both LrSO2 and RrSO2 increased to 77% and 87%, respectively, due to decreased cerebral oxygen demand by anesthetic effects. In this case, during CPB, both left and right rSO2 were above the critical threshold without taking optimization of oxygen balance into consideration. According to our experience, a decrease in bilateral rSO2 is not uncommon during the initiation and maintenance of CPB because of decreased cerebral perfusion as a consequence of non-physiological arterial flow and lower perfusion pressure. If there is any profound reduction in rSO2 at this stage, the depth of anesthesia, body temperature, CPB pump flow, CPB circuitry and perfusion pressure of CPB should be closely monitored.
After clamping of the proximal aortic root, distal descending aorta and arch vessels for antegrade cerebral perfusion, the LrSO2 and RrSO2 decreased to 61% and 70%, respectively. At this juncture, the retrograde RSA cannulation wholly provided crucial bilateral antegrade cerebral perfusion. Although the LrSO2 and RrSO2 were decreased by 13% and 15%, respectively, from the baseline value, both were still above the critical threshold, indicating that cerebral perfusion was adequate. The well functioning RSA cannulation and appropriate flow were important to maintenance of rSO2.
After declamping the LCCA following its anastomosis with the graft, the left cerebral hemisphere was supplied by the graft while the right cerebral hemisphere was still supplied by RSA antegrade perfusion. Both left and right rSO2 were proportionally increased without any acute or intense increase or decrease. LrSO2 was increased from 61% to 69%. We were of the opinion that this increase might not only indicate a patent LCCA anastomosis with fluent flow, but also signified the absence of cerebral hyperperfusion via the newly created LCCA. Meanwhile, RrSO2 was also proportionally increased from 70% to 74%, possibly due to the fact that RSA antegrade perfusion needed only to supply the right cerebral hemisphere without a share in supplying the left cerebral hemisphere. With subsequent anastomosis of the IA with the graft and declamping of the IA, RrSO2 was increased to 80 points without exceeding the baseline value. This might be due to greater blood flow from a larger-caliber IA and branch graft compared with RSA cannulation. In these stages, anastomoses and patency of these branches were crucial in maintaining adequate bilateral cerebral perfusion.
After smooth weaning from CPB, recovery of physiologic pulsatile cerebral perfusion was the reason why the LrSO2 and RrSO2 increased to 82% and 85%, respectively. At the end of surgery, LrSO2 and RrSO2 decreased to 73% and 74%, respectively, because of increased cerebral oxygen demand subjected to lightening of the anesthesia level.
In conclusion, cerebral oximetry is a documented comprehensive means of monitoring global brain perfusion.16 During complicated aortic arch surgery with concomitant branch reconstruction, cerebral oximetry can provide continuous information about cerebral oxygen balance.17 Before the operation, we should not only explain the risk carefully to the patient and family, but also prepare all necessary anesthetic facilities including the cerebral oximeter. During operation, adjustment of the anesthesia plan according to rSO2 monitoring could allow us to minimize the likelihood of the brain being subjected to inadequate oxygen supply.18 Here, we have described the serial changes in continuous cerebral oximetry in a patient undergoing complicated cardiovascular surgery involving interference of cerebral perfusion, and discuss the possible corresponding mechanisms in each stage of the surgical procedure. There were specific factors responsible for the changing rSO2 in individual stages of the operation. Therefore, one should modulate the possible influential factors so as to maintain the rSO2 above the critical threshold if there is a severe reduction in rSO2. The serial changes in this case can serve as a rational model for judging the circumstances in complicated cardiovascular surgery.